Oceanography

HMSC GC 856 .0735 no. 74-4A of cop.2 OCEANOGRAPHY

PROGRESS REPORT Ecological Studies in Radioactivity in the Columbia River Estuary and Adjacent Pacific Ocean Norman H. Cutshall Principal Investigator

Complied and ediled by Kayla J. McMechsn

U.S. Atomic Energy Commission Contract AT(45-1)-2227, Task Agreement 12 OREGON STATE UNIVERSITY RLO-2227-T1243 Reference 74-4 1 July 1972 through 31 March 1974

MARILYN"TT' G1 11t9RARY ii.:NCE CENTER

RE( QN SikiU i' ,. ?S TY

NEWPORT,Q GO 97365 117LI11

ECOLOGICAL STUDIES OF RADIOACTIVITY IN THE COLUMBIA

RIVER ESTUARY AND ADJACENT PACIFIC OCEAN

(USAEC Contract AT(45-1)-2227, Task Agreement12)

PROGRESS REPORT 1 July 1972 through 31 March 1974 RLO-2227-T12-43

Compiled and Edited by Karla J. McMechan

Submitted to Division of Environmental and Biomedical Research U.S. Atomic Energy Commission

NormanH.Cutshall Principal Investigator

Andrew G.Carey,Jr. Robert L. Holton WilliamG.Pearcy William C. Renfro Co-Investigators

School of Oceanography Oregon State University Corvallis, Oregon 97331

Reference74-4 John V. Byrne April 1974 Dean

MA ILYN PJ1TSGINLIBRARY EtD ° J NE SCIENCECENTER L UNIVERSITY OREG 97365 ii

ACKNOWLEDGMENTS

A major expense in oceanographic research is "time at sea."Operations on the R/VYAQUINA,R/V CAYUSE, R/VPAIUTE,and R/V SACAJEWEA were funded by severalagencies,with the bulk coming from the National Science Founda- tion and Office of NavalResearch. Certain special cruises of radiochemical or radioecological import were funded by the Atomic Energy Commission, as was much of the equipment for radioanalysis and stable element analysis. We gratefully acknowledge the role of these agencies in support of the research reported in the following pages. We also wish to express our thanks to the numerous students and staff who contributed to the preparation of this progress report.

NOTICE

The progress report that follows includes research results ranging from unproved ideas to scientific papers published during the tenure of this contract. The end of the contract year finds several facets of our work in various states ofpreparation,therefore the reader is cautioned that all except the published papers are subject to revision. iii

PRINCIPAL INVESTIGATOR

NormanH. Cutshall, Ph.D. Research Associate

CO-INVESTIGATORS

AndrewG. Carey,Jr., Ph.D. Assistant Professor

RobertL.Holton, Ph.D. Research Associate James E. McCauley, Ph.D. Associate Professor William G. Pearcy, Ph.D. Professor

WilliamC.Renfro, Ph.D. Assistant Professor

STAFF

NormanD.Farrow, B.S. Instrument Technician I. Lauren Larsen, M.S. Research Associate

EarlE.Krygier, M.S. Research Assistant R. Eugene Ruff, B.S. Research Assistant

KarlaJ.McMechan, M.A. Research Assistant

JeromeW.Wagner, B.A. Research Assistant

STUDENTS

DavidW.Evans, M.S. Graduate Assistant Priscilla J. Harney, M.S. Graduate Assistant Vernon G. Johnson, M.S. Graduate Assistant

HaroldL.Longaker, M.S. Graduate Assistant Janakiram R. Naidu, M.S. Graduate Assistant

KatsuoA.Nishikawa, B.S. Graduate Assistant

WalterH.Pearson, M.S. Graduate Assistant

BruceE.Rieman, B.S. Graduate Assistant

HenryA.Vanderploeg, Ph.D. Graduate Assistant V

TABLE OF CONTENTS

INTRODUCTION...... 1

STUDENT PARTICIPATION ...... 15

ABSTRACTS OF THESES ...... 19 The Dynamics of Zinc-65 in Benthic Fishesand Their Prey Off Oregon--Vanderploeg...... 19 Effects of Ocean Water on the Soluble-Suspended Distributionof Columbia River Radionuclides--Evans ...... 21 Loss of Zinc-65 and Manganese-54 From the Fresh Water Mollusc Anodonta--Harney ...... 22 Growth Response of a Marine Phytoplankton Cbccolithus huxleyi,

to Various Chemical Forms of Cobalt--Longaker ...... 23

MEETINGS ATTENDED ...... 25

PAPERS PRESENTED...... 27

RESEARCH IN PROGRESS...... 35

Post-Shutdown Radionuclide Transport Study--Johnsonand Cutshall 35

Effects of Seawater on MetalChemistry--Cutshall,Evans, Wagner and Farrow ...... 37 Cycling and Decline of Radioactivity in Columbia RiverEstuary

and in Alder Slough Biota--Renfro and Farrow ...... 38 Variations of Zinc-65 in Zooplankton Relativeto the Columbia

River Plume--Rieman and Pearcy ...... 39 Changes in the Radioactivity and the Specific Activity of Zinc-65 inEpipelagic,Mesopelagic and Abyssobenthic Animals Off Oregon--Pearcy ...... 39

Carbon-14 Content of Deep-Sea Animals--Pearcy and Stuiver... 40

Spatial Patterns of Zinc-65 Specific Activity in Pelagicand Benthic Food Webs on Oregon's Continental Shelf, July 1970-- Vanderploeg and Pearcy ...... 40

Biology and Systematics of Deep-SeaAnimals Off Oregon-- Stein and Pearcy ...... 41 vi

Description and Biology of a New Species of Bentheogennema (Decapoda, Aristaeinae)From the Northeastern Pacific-- Krygier and Wasmer ...... 41 Isotope Dilution of Zinc-65 in Oregon Coastal Waters-- Cutshalland Larsen ...... 42 Zinc-65 and Iron-55 Turnover in Columbia River Fishes-- Cutshalland Jennings ...... 42 Loss of Zinc-65 Specific Activity in Mussles (Mytilus ealifornianus) From the Oregon Coast Following Reactor

Shutdown--Larsen and Wagner ...... 42

Zinc,Radioactive Zinc(65Zn),Cadium and Mercury in the PacificHake,Merluccius productus(Ayres),Off the West

Coast of U.S.--Naidu and Cutshall ...... 43 Large Batch Culture Experiments in Radioisotope Kinetics--

Larsen,Holton and Cutshall ...... 44

Natural Radionuclide Indicators of DenseMineral Deposits

Along the Pacific Northwest Coastline--Scheidt and Cutshall . 44

Benthos:Benthic Ecology and Radioecology--Carey ...... 45

RESEARCH COMPLETED ...... 47 Sound-Scattering Layers in the Northeastern Pacific-- Donaldson and Pearcy ...... 47

Radioactivity in Juvenile ColumbiaRiverSalmon: A Model to Distinguish Differencesin Movement and FeedingHabits-- Romberg and Renfro...... 54

SpecificActivityof Iron-55 in Pacific Salmon--Jennings and Osterberg ...... 60 Effect of Gamma Irradiation on the Reproductive Performance of Artemia as Determined by Individual Pair Matings--Holton,

Osterberg and Forster ...... 66 Effect of Gamma Irradiation on the Maintenance of Population Size in the BrineShrimp,Artemia--Holton, Osterberg and Forster . 73

Zinc-65 and DDT Residues in Albacore Tuna Off Oregon in1969-- Pearcy and Claeys ...... 81 New Records for Four Deep-Sea Shrimps from the Northeastern Pacific-Wasmer...... 89 vii

Zinc-65 in Oregon-WashingtonContinental Shelf Sediments-- Cutshall, Renfro, Evans and Johnson ...... 94 Zinc-65 Specific Activity inMytilus californianusTissues-- Larsen, Renfro and Cutshall ...... 103 Seasonal Radionuclide Inventories in Alder Slough,an Ecosystem in the Columbia RiverEstuary--Renfro ...... 108 Rate of Zinc Uptake byDover Sole in the NorthwestPacific VOcean: Preliminary Modeland Anaylsis--Vanderploeg ...... 117 Losses of 65Zn to InorganicSurfaces in a Marine Algal Nutrient

Medium--Tomlinson and Renfro ...... 126 Food Sources ofSublittoral,Bathyal and Abyssal Asteroidsin the Northwest PacificOcean--Carey ...... 131 A Redescription of Petalidium suspiriosum Burkenroad,1937 (Decapoda: Natantia)--Wasmer ...... 144

A Technique for the Estimationof Indices of Refraction of Marine Phytoplankters--Carder, Tomlinson and Beardsley ...... 157 Growth of a Sea Urchin, Allocentrotus fragilis,off the Oregon Coast--Sumich and McCauley ...... 164

Estimating Precision forthe Method of Standard Additions-- Larsen, Hartmann andWagner ...... 176

First Records ofBathysaurus mollis(Pisces: Bathysauridae) From the Northeast Pacific Ocean--Stein and Butler ...... 179

Swimbladder Morphologyand Specific Gravity of Myctophids Oregon--Butler and Pearcy off ...... 181 Zinc-65 Specific Activities from Oregon and WashingtonContinental Shelf Sediments and BenthicInvertebrate Fauna--Carey and Cutshall ...... 189

: Radioecology of Benthic Fishes Off Oregon--Pearcyand Vanderploeg...... 208 Effects of Ocean Wateron the Soluble-Suspended Distribution Columbia River Radionuclides--Evans of and Cutshall ...... 225 Calcium-Magnesium Ratiosin the Test Plates of AZZocentrotus fragilis--Sumich and McCauley ...... 241 viii

A Systematic Review of the Rattail Fishes (Macrouridae: Gadiformes)from Oregon and Adjacent Waters--Iwamoto and Stein ...... 246

Turnover of 65Zn in Oysters--Cutshall ...... 247 Linear Instrument Calibration with Statistical Application-- Larsen and Wagner ...... 260

Food Habits of Deep-Sea Macrourid Fishes off the Oregon Coast- Pearcy and Ambler...... 274

Heavy Metals in Estuaries and the Coastal Zone--Cutshall ... 293

PUBLICATIONS...... 299 INTRODUCTION

This report presents progress accomplished during the 20-month period from 1 July 1972 through 31 March 1974 as part of the program "Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent Pacific Ocean". This program is a continuing study supported by the Division of Biomedical and Environmental Research of the U.S. Atomic Energy Commission under ContractAT(45-1)-2227,Task Agreement 12 with Oregon State University.

The report format is one which has evolved in recentyears. In this introductory section we attempt to describe the history and goals of the program and to identify the major physical facilities which are dedicated primarily to the program. This presentation is followed by identification of students associatedwith theprogram and then the results of recently completed thesisresearch. Next,staff participation in meetings is recounted. The various research projects which are underway but for which final papers have not been written are thendescribed. In the Research in Progress section we indicate the rationale and approach to specific research projects and keep the reporting of data and conclusions to a minimum. The largest section of the report is ResearchCompleted,wherein we reproduce manuscripts which have been submitted for publication and reprints of paperspublished. Papers included as reprints may have been printed in earlier progress reports asmanuscripts. Although a certain amount of repetition thereforeoccurs,in many cases the text of published papers varies from the text originallysubmitted. Finally,we include a bibliography of publications from the program.

Marine Radioecology at Oregon State University (updated from 1971 Progress Report) Chronology Marine radioecology at OSU began in the early 1960's.In early 1961 Charles Osterberg demonstrated that neutron-induced radionuclides from the Hanford plant could be readily measured in marine organisms collected off the Oregoncoast. Three Atomic Energy Commission research contracts were sooninitiated: ContractAT(45-1)-1726,Species Composition and Distribu- tion of Marine Nekton in the Pacific Ocean offOregon,withDr.William Pearcy as principalinvestigator,began inSeptember, 1961;in June, 1962, Radioanalysis of oceanic organisms in the Pacific Ocean offOregon,Contract AT(45-1)-1750,was initiated withDr.Wayne V. Burt as principal investigator and Charles Osterberg asco-investigator;ContractAT(45-1)-1758,Ecological and Radioecological Study of the Benthos in the Pacific Ocean off Oregon, began inFebruary, 1963,with Dr. AndrewCarey,Jr., as principal investiga-

1 2 for and Drs. McCauley and Osterberg as co-investigators. These three studies comprised a broad marine ecological program with emphasis upon artificial radionuclide cycling. In November, 1964,they were officially combined under the -1750 Contract number with Dr. Osterberg as principal investigator. In November, 1970, the contract number was changed to AT(45-1)-2227, Task Agreement 12, Since June, 1967, Dr. Osterberg has been on leave from OSU. During this period the program has actively continued with Drs. Forster (1967, 1968), Renfro (1969) and Cutshall (1970 to present) as principal investigators.

Philosophy

The OSU radioecology program was begun and has continued with three basic tenets:

1. In order to understand the impact that artificial radioactivity may have upon an ecosystem, it is necessary to first have a good working knowledge of the mechanics of that ecosystem.

2. Direct determination of radionuclide distribution in an ecosystem is needed in order to test our understanding of the system.

3. Artificialradionuclides themselves provide a most useful tool for learningecosystem mechanics. Thus, the Columbia River System has been considered a "unique natural laboratory for radioecological studies" (1964 proposal).More conventional ecological study of species abundance, distribution, their interactions with one another and with the physico-chemical environment has been facilitated by the presence of radioactive tracers.These principles continue to guide our research. Whereas the original primary focus was upon the Columbia River System, it is now turned to coastal marine and estuarine areas in general.We are also engaged in smaller scale studies on deep ocean and laboratory problems. We continue to be principally concerned with the physical and biological cycling of trace materials in coastal ecosystems, with emphasis upon materials introduced by man's activities and which potentially may affect the biotic community.

Results of the Program Publications resulting from the program have covered a wide range of topics.Together with the papers of other AEC-supported researchers at the University of Washington, the laboratories at Hanford and the US Geological Survey in Portland, Oregon, they embody the great majority of knowledge about the Columbia River System.The list of publications included in this report includes titles on virtually every aspect of the system.In addition, several hundred other School of Oceanography papers provide a considerable breadth of related knowledge. 3

Students have provided a considerable resource of manpower and ideas to the program. Fifty graduate degrees have been awarded to students in or closely associated with the radioecology program. Of these, 34 were Master of Science or Master of Arts degrees and 16 were Doctor of Philosophy degrees. Two individuals have received 2 degrees each. The majority of graduates are presently employed in oceanographic research with the primary employers being universities and the Federal Government.

Present and Future When the final production reactor was permanently shut down in January 1971,we faced adilemma. On the one hand an important (perhaps the most important)phase of the radioactive history of the Columbia River System wasbegun. The rates at which various components of the system discharged their radionuclide burden could be directlymeasured. Clearance times for natural systems will be of vitalimportance,should a major radionuclide release everoccur. On the other hand intensive study of rapidly decaying radionuclides was clearly a terminalexercise. It seemed imperative to us that we not only increase both the detail and scope of our "Hanford radionuclide"studies,but also that we simultaneously develop other lines of research effort which had a more promisingfuture. Our analytical facilities for radionuclide and metal analyses were obsolete and failing rapidly.

The past three years have been transitional building years. New projects have been started in our Radioecology program. Related programs have been undertaken by our group. By the end of the present contract year, June 30, all major instruments will have been updated to state-of-the-art. Studies of the decline of radionuclides in the Columbia River System are nearing completion.

Our principal long term goalsare to assessand predict the impact that man's energy-producing efforts will have on coastal ecosystems. In order to realize these goals, we will pursue a broad and diverse set of detail-level studies aimed at understanding the systems themselves. We will emphasize direct analysis of the suspected impact-producing byproducts in the natural environment. We will carefully examine the results of these analyses in order to detect flaws in our conception of how the systems function.

Related Programs

Three programs closely related to the AEC program,have been underway during the past year.

Dr. Cutshall is Principal Investigator on "Effects of Ocean Water on the Physico-Chemical Form of Heavy Metals", supported by the office of International Decade of Ocean Exploration of the National Science Foundation. The project seeks to detect changes in the chemical and physical form of zinc, copper, manganese, cadmium, lead and mercury during their transport through estuaries. This research is based directly upon earlier AEC studies of Hanford radionuclides. 4

Drs. Holton and Cutshall are co-investigators withDr.Slotta of Ocean Engineering (OSU) on"Physical,Chemical and Biological Studies of Young's Bay" supported by American Metals Climax Corporation(AMAX). The program is a broad baseline study of the Young's Bay portion of the Columbia River Estuary. AMAX intends to build an aluminum reduction plant near the bay and seeks information on ecological conditions prior to beginning construction. These studies are being closely integrated with our AEC studies in the estuary and particularly Alder Slough, which is adjacent to Young's Bay. Dr. Holton is Principal Investigator on "The Development of Methods for Studying Physical and Biological Processes in the Nearshore Zone on the Pacific Coast of the UnitedStates,"supported by the Joint Power Planning Council.This program involves research in the surf zone and the area immediatelybeyond,with emphasis upon possible effects of coastal thermal electric power plants on biota.

Facilities Major improvements have been made in our analytical laboratories in the pastyear. These are expected to be completed by September, 1974. Chemical Analysis Laboratory The backbone of our chemical laboratory has been the Perkin-Elmer 303 atomic absorption spectrophotometer purchased with AEC funds in 1964. This basic instrument has now been supplemented by a Varian AA-5R system provided by the National ScienceFoundation. In addition we have recently added a Varian Model #635 UV-visible spectrophotometer for our AMAX program. Atomic Absorption Spectrophotometer System (See Figure 1).The AA-5R atomic absorption spectrophotometer is a single beam instrument utilizing a modular component type assembly.The solid state tuned amplifier is of a pulse modulated, phase-locked design providing an exceptionally stable electronic system.The four quadrant, hollow cathode lamp turret assembly, burner chamber and monochrometer are mounted on an optical rail providing stability and system flexibility.Both flame absorption and flame emission capability are selectable.

Readout inabsorbance,concentration or per cent transmittance is available by meter on a four digit nixie-tube display. Two additional forms of outputs are utilized.They are a multi-speed, multi-scale, ten-inch wide strip-chart recorder and a model ASR33 teletype unit interfaced to thesystem. The teletype provides typed copy output and/or punched paper tape (ASCII code), so that operator error in recording data iseliminated. An interface between the teletype and the atomic absorption spectrophotometer (AAS) provides a selection of either a single reading preceeded by a four digit identification number or multiple of ten readings preceeded by a four digit identificationnumber. The punched paper tape output is used as the data file for computer analysis described later. Ln Figure 1.Atomic Absorption Spectrophotometer System. 6

Other features unique to this system includes a two channel digital curve corrector, a fifty-place automatic sample changer and a model 63 carbon rod flameless atomizer system. The digital curve corrector provides two variable channels which are used to compensate for deviations from Beer's Law when needed. The automatic sample changer provides practically attendant-free operation, since it is interfaced with the AAS and teletype system and also controls the automatic baseline corrector. Two adjustable timers on the automatic sample changer can be set to control the duration of sample uptake and of rinsing between samples. The automatic changer has a capacity of fifty samples.

The carbon rod flameless atomizer consists of a small carbon tube (about 5 mm o.d. by 10 mm in length) supported by two carbon electrodes connected to a programmed low voltage, high current power supply. Three stages of programming allow the sample introduced into the carbon tube to be dried, ashed and atomized automatically. The carbon rod flameless atomizer produces a dense population of ground state atoms in a confined absorbance zone with an increased residence time within that zone compared to flame atomization. Because of the increased atom population and residence time, sensitivity and detection limits are improved two to three orders in magnitude for many metals. In addition the sample volume requirements are greatly reduced (typically .005 ml per determination), interferences are more readily controllable and there are some applications for direct analysis of solid samples.

We now havehollow cathodelamps for 26elements: Ag, Al, Au, Ca, Cd, Co, Cr, Cs, Cu,Eu, Fe, H2, Hg, K, Mg, Mn,Mo, Na, Ni, Pb, Pt,Ru, Sc, Si, Zn, and Zr.

Computer Programs for AAS.Five computer programs have been prepared for AAS data reduction and error analysis.They are: "PRINPT", "AA5TTY", "MOA" and "CARBON".Each program was written in FORTRAN IV for a PDP-15 computer for a specific task dependent on the method of analysis.Output from the AAS teletype, in the form of punched paper tape (ASCII code), contains the raw data file which is then read by the computer's optical tape-reader terminal.The computer output, after computation, is available in three forms:magnetic tape (DEC tape), punched paper tape (ASCII code) and printed page (high speed line printer (HELP) or teletype printer). The data file is transferred to magnetic tape for storage in most cases, The reduced data and error analysis are recorded on eleven by fifteen inch computer sheets or on eight and one half by eleven inch teletype sheets. Program "PRINPT" was developed primarily for a rapid error analysis of the AAS data.The punched paper tape from the AAS teletype unit is read into the computer and the output printed by the computer's HSLP, which lists the sample identification number, arithmetic mean (ten replicate measurements), standard deviation, relative per cent standard deviation, standard error, per cent standard error, 95 per cent confidence level and the per cent error at the 95 per cent confidence level.The program does not compute a corrected sample mean, i.e. it does not correct for blanks, intercept or slope of the standards, 7

Program "AA5TTY", written to interpret the raw data file from the teletype output of the AAS unit which contains the instrumentresponse to the standards and "unknowns", computes a "best fit" linear calibration line and determines a corrected sample concentration witherror analysis for each "unknown". The input files to the computer consist of two parts. The first is the raw data file on punched paper tape and the second isan identification file, also on punched paper tape, which contains the sample identification, the quantity of sample that was digestedor diluted, the volume it was diluted to and other dilution or concentration factors. The computer then analyzes all for the standards that were measured and performs a "Least Squares Linear Regression" analysis.1The computer prints out the error analysis for the linear regression model and then computes the sample "unknown" using the regression parameters suchas slope and intercept values. The HSLP lists the sample number, sample identification (i.e. species name, location and date of collection, etc.), sample concentration, standard error, per cent standard error, arithmetic mean of the instrument response corrected for slope and intercept, sample weight and sample volume.

Program "MOA" (methods of addition) was specifically tailored to a data reduction system for analyzing trace metal concentrations insea water by a solvent extraction technique. Two files are merged during computation by the computer. The first file is the raw data file for the AAS teletype (punched paper tape), the second file contains information about the location of the sample, standards and amount of spiked metal concentration added to each sample. Through a brief conversational exchange, using the computer's teletype, additional information is transferred, such as sample identification, date collected, output concentration form and concentration factor. The computer analyzes the samples and standards using a "Linear Regression Least Squares" model, then estimates the extrapolated value for each set of samples, along with an error analysis based on the method described by Larsen, Hartman and Wagner (1973).2 The output lists parameters for the linear regression analysis of the standards, sample identification, correlation coefficient for each set of samples, net partsper million, y- intercept, slope of each set of samples, sample variance, number of data pairs, pooled standard error of regression, estimated precision for the difference (standards and samples), degrees of freedom, confidence interval (in concentration units) at the 95 per cent confidence level,per cent confidence interval at the 95 per cent confidence level and sample concentration (in micrograms metal per liter).

Program "CARBON" is used to evaluate the data obtained from the carbon rod flameless atomizer technique. Sample and standard data are transferred to punched paper tape, then to magnetic tape for computer analysis and storage. Another file is generated on punched paper tape which identifies the sample and standards along with sample collection and processing information, such as salinity, concentration factors, and sample volume, etc.

1 Bevington, P.R. 1969. Data Reduction and Error Analysis for the Physical Sciences. McGraw-Hill Book Co., New York. 336 p. 2 Larsen, I.L., N.A. Hartman, andJ.J. Wagner. Estimating precision for the method of standard additions. Analytical Chemistry 45:1511-1513. 8

The computer analyzes the standards using the aforementioned linearregres- sion model and outputs the regression parameters of the standardssuch as slope, intercept, variance, correlation coefficient and otherassociated error terms. It then analyzes the samples and determines the sampleconcen- tration and resulting error terms. Output, using the HSLP, includes sample identification, information concerning sample processing,concentration of the sample, standard deviation, standarderror, uncertainty (in terms of the sample concentration units) at the 95per cent confidence level, per cent uncertainty at the 95 per cent confidence level, peak heightresponse of the sample and total quantity of metal extracted.

Counting Laboratory

We are now operating two gamma-ray spectrometers utilizing 5"x 5" NaI(TI) well detectors and 512-channel multichannelanalyzers. A third analyzer is alternatively coupled to 3"x 3" or 3" x 4 1/2" solid detectors. We expect to use this analyzer in conjunction witha small effort in alpha spectrometry.

Bids have been received fora new low level gamma counting system which we expect to be in operation during the present contract year. This system will utilize a Ge(Li) primary detector, the signalsfrom which are routed according to a surrounding NaI(TI) shield. Data will be acquired and analyzed using a hybrid 8192-channel analyzer and 16K, 16-bit processor.

Specifications were written from specifications originallywritten by W.H. Zimmer of Atlantic Richfield Hanford Company. Good qualities of the specifications should be attributed to him. Others are our own.

Specifications for a Low Level Anticoincidence-Shielded, Lithium-Drifted Germanium Gamma Ray Spectrometer System

A. System Specification

Al. "System" means the entire set of components specified below, inter- connected and operating and includes all interfacing hardware and software. Lead shielding and apparatus for moving the detectors is not part of the system.

A2.The systemwill be installed, checked anddemonstrated to operate in the purchaser's laboratory in Corvallis.

A3. Vendor will assume complete system responsibility for a period of one year following installation. Any repair costs resulting other than from misuse shall be borne by the vendor. A4.Delivery time shall not exceed 90 days after acceptance of the bid. 9

AS. Components include: I. Detectors and front end electronics a. Ge (Li) Detector b.NaI(T1) Annulus c.NaI(Ti) Plug d. Power supplies for above e. Amplifier, logic and routing devices so that Ge(Li) events coincident and not coincident with NaI events can be digi- tized and recorded in separate portions of memory II. Analyzer and Controller a. 100 mHz analog to digital converter b. Analyzer c. Digital controller III. Peripherals a. Magnetic tape and control (1/2", 25 ips) b. Interface to Tally 420 and Tally 424 tape perforater and reader. Includes software to read and punch Nuclear Data (series 130AT) format. c. Hewlett-Packard Model 7004B, 11" x 17" Point Plotter d. Hewlett-Packard Model 1300A 13" Cathode Ray Tube

B. Analyzer Specifications

Data Acquisition and Storage Memory

B1. This memory must contain 8192 channels of 24 bit ferritecore memory.

B2. Memory cycle time shall be.equal to4.0 microseconds.

B3. Memory must be programmable to add or subtract data with a toggle switch.

B4.Memorymust be subdividable into 16 subgroups by, front panel rotary switches or external routing signals.

B5. Data must be transferable from one subgroup to any otherequal size subgroup without use of computer control.

B6. The memory unit must be capable of pulse height analysis or multichannel scaling with internal programming or with computer control.

B7. The memory must be capable of multiplex operation to accept pulses from two ADC's into separate halves,

B8.Basicanalyzer functions must be controllable with externalcomputer through rear panel signals connector.

B9. MCSdwell times from 10 microseconds through0.9 seconds must be included in built-in crystal controlled clock.

B10. Presetlive or clock times must be built in and switch selectable. A separate live time must be storable in the first channel of every memorysubgroup.

Bll. A 5" CRT must be built into the memory unit with times 5 horizontal and vertical expansion and position controls.

B12. Dual parameter analysis must be possible with the memory unit including isometric, contour and profile displays. 10

B13. Every10th or 100thchannel must be intensified for channel identification and selected by switch control. B14. Selected readout of an arbitrary memory size must be included with intensification markers for both X or Z levels. B15. Memory must include full analog output signals for use with an X-Y plotter. B16. The memory must include its own power supply to operate on 110 or 220V/50 or 60 Hz power.Maximum power consumption must be 120 watts. B17. A switch selectable program must be wired in for manual, automatic stop and automatic stop, destruct and recycle operation with or without computer control. B18. The memory must be configured for two parameter data acquisition in group sizes of 1 x 8192 through 64 x 128 for an 8192 channel memory.

B19. A live static or normal static display shall be switch selectable. B20. A full-scale X-Y calibration button must be included for calibration of CRT display or X-Y recorder. B21. The memory unit must be mountable in a standard 19" rack and have maximumdimensions of l2"H x 16"W x 22"D.

Memoryto Digital ControllerInterface B22. The interface shall provide for full bidirectional control and transfer of data to and from the storage memory unit and 16 bit digital controller. B23. This interface shall include provisions for addition of other peripherals in the future with additional circuit boards. B24. The interface shall include its own power supply and power switch. B25. It shall be packaged in a standard 19" rack mountable cabinet with a maximum height of 8-3/4". Digital Controller Purchase Specifications B26. The digital controller must use a 16 bit word format with core sizes of 16K words, expandable to 32K words. B27. Memory cycle time must be 1.2 microseconds or less.

B28. Complete control panel must be supplied with the controller.

B29. A real time clock shall be included.

B30. An ASR33 Teletype and interface to the controller must be included for control and readout.

B31. The controller must include 4 independent accumulators.

B32. A full assembler and editor must be included with the controller. 11

B33. A one week school in operation and programming execution for this controller must be included in the quoted price.

B34. Controller must contain direct memory access channel, auto restart circuit, automatic program load and power monitor.

B35. The controller must contain 16 priority interrupt levels.

C. Detector Specifications

Lithium Drifted Germanium Detector

Cl. The detector will be a closed end coaxial Ge(Li) detector.

C2. Detector system is specified for purposes of resolution to mean the detector with the preamplifier, amplifier and base line restorer supplied as part of the total system.

C3. Detector system resolution, full width at half maximum (FWHM) for the 1332 keV peak of 60Co shall be equal to or less than 2.0 keV.

C4. Detector system resolution full width at tenth maximum (FWTH) for the 1332 keV peak of 60Co shall be equal to or less than 3.9 keV.

C5. Detector system resolution, FWHM for the 122 key peak of 57Co shall be equal to or less than 0.95 keV.

C6. The detector system relative efficiency shall be at least fifteen percent. Relative efficiency is defined as the ratio of the 1332 keV peak area of 60Co determined from a point source twenty-five centimeters from the detector face for the Ge(Li) detector compared to a 3" x 3" Na(Tl) detector expressed as a percentage.

C7. The detector system peak to Compton ratio for the 1332 peak of60Co shall equal or exceed 38:1. Peak to Compton ratio shall be measured as a ratio between the peak height of 60Co at 1332 and its associated average Compton edge.

C8. The dead core diameter in the detector shall be less than fourteen millimeters.

C9. The high voltage filter network shall be in preamplifier case, external to the cryostat.

C10. The high voltage (bias) feed through from the cryostat to the preamplifier case shall be separate and distinct from the pass through for all other signals.

Cll. The diode shall be mounted with the closed coaxial end closest to the cryostat "window" and within five millimeters of it.

C12. The diode mounting shallbe magnesium of the minimum mass consistent with structural strength.

C13. All components internal to the cryostat shall be selected first on the basis of minimum contribution to radioactive background and second on the basis of minimum mass.

C14. The detector end cap will be 8 inches long and will be compatible with the sodium iodide annulus specifiedelsewhere herein. C15. The Dewar shall have a liquid nitrogen capacity of thirty liters or greater. 12

Anticoincidence Shield

C16. The annulus dimensions in its can shall be a right cylinder equal to or greater than nine and three eighths inches'in diameter, equal to or greater than twleve inches long.

C17. The inside diameter shall accomodatea standard HarshawIntegral Line 3 x 3"NaI(Tl) detector withequal or lessthan 0.05inches radius clearance. C18. The loading factor of NaI(Tl) in the annulus shall equal or exceed 99% excluding reflector and enclosing can. C19. The exterior can shall be low radioactivity aluminum. C20. Resolution shall equal or be less than fifty percent. C21. There shall be four 3" multiplier phototubes all mounted on one cylindrical end of the annulus. C22. The multiplier phototubes shall be individually attached in housings which will permit customer replacement of tubes without damage to the detector and without recourse to specialized tools or a dry box. C23. Each multiplier phototube shall be fitted with a low noise, non- phenolic base and a magnetic shield.Each base shall contain the following: A standard resistor voltage divider network A gain control with operating knob Focuscontrol which shall be set and locked by the supplier An anode signal lead through jack, Amphenol Connector #5116-058350 or approved equal A high voltage lead through BNC type jack

C24. Support for the entire annulus assembly shall be from the flange at the lower, phototube end, of the annulus.Predrill at least four holes in the flange to position and secure the assembly.

C25. The annulus cladding can be either single or double flange construction.

Plug Detector

C26. A standard Harshaw Integral Line 3 x 3" NaI(Tl) detector will be supplied.

D.Detector Electronics

Dl.Providea 5000 VoltBias Supply for the Ge(Li) detector.

D2. Provide a 0-3000 Volt Power Supply forthe NaI(Tl) detectors. D3. Provide suitable preamplifiers, amplifiers,logic and signal con- ditioning hardware for presentationof coincident and noncoincident events in the Ge(Li) detector to the analyzermemory. D4.Provide over-voltage protection foreach highvoltage power supply (Dl. and D2. above) such that, ifthe line voltage drops below 90V AC for 0.5 second or moreor the line voltage fails completely, the power supply or highvoltage section will beelectrically disabled requiring manual reset toresume normal operation. 13

E.Software Specifications El.Providethe followingmachine language software: a. For all input/output components. b.For control of total systemor analyzer section or processor section from Teletype keyboard. c. For identification of spectralpeaks and quantification basedon referenceconstants, including efficiency corrections,decay corrections and statistical precisionestimates. E2. Provide a BASIC compiler.

Other Facilities We are planning to modify our chemical preparation laboratory to include an isolated, positive pressure room with a laminar flow hood facility.This area will be used for processing of environmental samples for ultra-low level analyses where contamination has heretofore prohibited accuracy.Tentative plans have been drawn.We expect to have the laboratory on line late in 1974. A marine culture room for intermediate level (millicurie) radionuclide turnover studies has been outfitted within the OSU Radiation Center building. Running water aquaria and large volume (350 gallon) static tanks have been installed (Figure 2) and isotope experiments, discussed under Research in Progress, are scheduled to begin in Spring 1974. 14

20' 'Ir

FLOWING LARGE VOLUME TANKS - 1000 liters0

WATER

000N

AQUARIA

Figure 2.Configuration of AquariumRoom A144,0SU Radiation Center. 15

STUDENT PARTICIPATION

Students contribute substantially to our research program. During the past year one student has completed his doctorate and four have completed the master's. Abstracts of their theses are printed in the following section. Thesis advisors are named in parentheses.

Degrees Completed

Henry Alfred Vanderploeg, Ph.D. Dr. Vanderploeg presented his thesis "The Dynamics of 65Zn in Benthic Fishes and Their Prey off Oregon" in November, 1972.He is presently on the Environmental Sciences Division staff of Oak Ridge National Laboratory. (Pearcy)

David William Evans, M.S. Mr. Evans completed his thesis research in June 1972 and worked with Dave Robertson and Richard Perkins at Battelle Northwest Laboratories as a Richland Graduate Fellow during the summer of 1972.Following presentation of his thesis in October, he began a two-year stint with the Atlantic Estuarine Fisheries Center of the National Oceanic and Atmospheric Adminis- tration in Beaufort, North Carolina.There he is working on his doctoral research in collaboration with Drs. Douglas Wolfe and Ford Cross.He is expected to return to OSU in Fall, 1974, for completion of requirements for the Ph.D. (Cutshall)

Priscilla Jeanne Harney, M.S. Ms. Harney has completed her M.S. degree in oceanography.Her thesis research was a study of the decline of radioactive zinc and manganese in a fresh water bivalve in the Columbia River system after the shut down of the production reactors at Hanford in 1971. The title of the thesis is, "Loss of Zinc-65 and Manganese-54 from the Freshwater Mollusc,Anodonta. She is now employed by the Oregon State Highway Department in Salem, Oregon. (Holton)

Harold L. Longaker, M.S.

Mr. Longaker has completed his M.S. degree in oceanography. His thesis research was a study of the effect of various forms of cobalt on the growth of a marine phytoplankton species. The thesis title was, "Growth Response of a Marine Phytoplankton,CoccoZithus huxZeyito various chemical forms of cobalt. Mr. Longaker has now entered a Ph.D. program at Oregon State Univer- sity. (Holton) 16

Ph.D. Candidates

David W. Evans Mr. Evans holds a B.S. degree in chemistry from UCLA and an M.S. in oceanography from OSU (see degrees completedsection). His doctoral research involves chemical reactions of transition metals in estuarine systems.

Vernon G. Johnson AEC Graduate Assistant

Mr. Johnson holds an M.S. degree in Oceanography from OSU.He spent four years at the National Reactor Testing Station in Arco, Idaho, and then returned to OSU to work toward the doctorate.His thesis research, which deals with radionuclide transport in the Columbia River following reactor shutdown, is nearing completion. (Cutshall)

Janakiram R. Naidu AEC Graduate Assistant Mr. Naidu received the M.S. from the University of Washington and then worked for six years with the India Atomic EnergyCommission. His thesis research involves comparison of65Zn,Zn, Cd and Hg in hake collected along the West Coast of the U.S. (Cutshall)

WalterH.Pearson IDOE Graduate Assistant Mr. Pearson received his M.S. degree at the University of Alaska in 1969. His thesis research involved the study of an intertidal marsh. He then spent two years in theservice. He joined our group in the fall of 1971. In the fall of 1972 he spent several weeks collecting samples for our program during a resurvey of the Bikini testsite. His thesis research is currentlyunderway. He is studying the effect of a pollutant, poly- chlorinatedbiphenyl,on the behavior of the crab, Hemigrapsus oregonensis. (Holton)

Master of Science Candidates

KatsuoA.Nishikawa Mexican Government Fellow

Mr. Nishikawa holds aB.S.degree from the Escuela Ciencias Marinas inEnsenada,B.C.,Mexico. Before joining us he was Director of the Institute of Marine Research at the University of BajaCalifornia. He has conducted a number of projects related to marine water quality, oil pollution and chlorinatedhydrocarbons. His thesis topic will concern metals in submarine thermal springs. (Cutshall) 17

Bruce E. Rieman AEC Graduate Assistant

Mr. Rieman started his Master's program in the fall of 1973. He completed analysis of our data on 65Zn specific activities in plankton and nekton collected on a series of cruises off the Oregon coast in 1969. Unfortunately he had to leave O.S.U. after fall term for non-academic reasons. (Pearcy) 19

ABSTRACTS OF THESES COMPLETED

THE DYNAMICS OF ZINC-65 IN BENTHIC FISHES AND THEIR PREY OFF OREGON*

Ph.D. Thesis by Henry Alfred Vanderploeg

The intra- and interspecific differences of 65Zn specific activity in benthic fishes on the continental shelf off Oregon during 1970-1971 are examined. The dynamics of 65Zn specific activity in the fishes are shown to be governed by the basic equation:

dS dt= a (F(t) - S) - SA

whereS = 65Zn specific activity of the fish (or predator), a = rate of input of Zn per body burden of Zn in the fish, F(t) = 65Zn specific activity of the prey, and X = physical decay constant of 65Zn.

This equation applies generally to any radionuclide accumulated through the food chain. Differences in S among fishes are thus related to differences in a and F(t). Using existing energy relations in fishes, it was hypothesized that a varies with the weight of the fish, W, and with growth:

a = K(AW-0.2 +1 W) r W

For a single location on the continental shelf off central Oregon, extensive time series of S and F(t) were obtained for different size classes of two species of benthic fishes. A numerical solution of the basic equation coupled to a least-squares gradient algorithm enabled calculation of a's from the time series of S and F(t). Usable timeseries wereobtained for a small flounder,Lyopsetta exilis.For L. exiZis size classeswith "average" weights of 22 and 35 g, the a's obtainedwere2.7 x 10-3/day and 2.6 x10-3/day, respectively. The a's ofL. exiliswere compared to theaderived from trout data that another worker obtained in the laboratory. The difference in a's betweenL. exiZisand the laboratory fish are roughlythe same aspredicted by the hypothetical a-W relationship, thus giving it tentative support.

* Major Professor: William G. Pearcy, Ph.D. 20

Although differences in S are expected according to the a-W relationship,these differences are small compared to those caused by the different 65Znspecific activities in differentprey. Specific activities of low trophic level pelagic prey were higher than those of high trophic level pelagic prey or of infauna.These trends in the prey were reflected in their predators, the fishes. The mechanisms behind geographical patterns in 65Zn specific activity in benthic fishes and their prey arediscussed;inparticular,these patterns are related to the 65Zn specific activities of sediment andwater. Near the mouth of the Columbia River which carries 65Zn from the Hanford reactors to theocean,infauna was seen to be as important as low trophic level pelagic prey in conveying 65Zn to the fishes.As distance increased from the river mouth,the importance of the infauna diminished relative to low trophic level pelagicprey. Evidence is presented that suggests that 65Zn was in a form more available to the food chain than stable Zn in seawater. The dynamics of 65Zn SA in the fishes are useful for understanding some aspects of thefishes' ecology. A theoretical framework is set up for determination of energy flow in free-living populations of fishes and other animals from their specific activitydynamics. Geographical patterns in 65Zn specific activity in the benthic fishes and their prey are shown useful for deducing the migratory habits of the fishes.Migration or its absence is suggested for some fishes for which no literature is available. 21

EFFECTS OF OCEAN WATER ON THE SOLUBLE-SUSPENDED DISTRIBUTION OF COLUMBIA RIVER RADIONUCLIDES* M.S. Thesis by David William Evans

The relationship of dissolved concentrations of Hanford radionuclides with salinity in the Columbia River estuary was interpreted in terms of the exchange of the radionuclides between dissolved and suspended particulate phases. Both 65Zn and 54Mn carried by the Columbia River were partially desorbed from suspended particulate matter upon mixing with ocean water in theestuary. Experiments in which ocean water was added to Columbia River water and to suspended particulate matter collected on filters confirmed the partial desorption of 65Zn and 54Mn from the particulate phase. The percentage of 65Zn and 54Mn desorbed varied with experimental approach but desorption of 65Zn seemed to lie in the range 15% to 45% and 54Mn in therange 30% to 60%. None of the experiments revealed any effect of salinity upon the soluble- suspended particle distribution of51Cr, 124Sbor46Sc. Dissolved concentrations of these nuclides varied inversely withsalinity. There was no evidence that any of the radionuclides studied was removed from solution byflocculation,precipitation or adsorption as the result of mixing with ocean water.

Ocean water contact partially removed 54Mn but not65Zn,46Sc or 60Co from Columbia River bottom sediments transferred to the marine environment. The inability of ocean water to desorb 65Zn from bottom sediment contrasts with its action with suspended particulate 65Zn.

* Major Professor: NormanH. Cutshall, Ph.D. 22

LOSS OFZINC-65 AND MANGANESE-54 FROM THE FRESH WATERMOLLUSCANODONTA* M.S. Thesis by Priscilla Jeanne Harney

Loss of 65Zn and 54Mn from the freshwater mollusc Anodonta was examined to determine possible radionuclide sources of these organisms, with sediments being of particular interest, and to make a field-laboratory comparison of effective radionuclide loss rates. A comparative field loss study was performed to determine if uptake of 65Znand 54Mn by Anodonta occurred after shutdown of the plutonium production reactor at Hanford, Washington.Loss rates of these two nuclides were measured in organisms transferred to the non-radioactive Willamette River and compared to loss rates found in organisms remaining in the Columbia River.Resulting 65Zn half-livesof 103 + 5 days and 135 + 12 days for transfer and in situ groups respectively indicated that significant uptake took place after shutdown.Periodic sacrifice collections of Anodonta from McNary Reservoir and the estuary further confirmed this finding.Manganese-54 uptake was less dramatic.Transfer and in situ comparisons were obscured by large individual variability.Sacrifice collections yielded ecological half-lives that were not significantly different from the pooled effective half-life. However, close examination of in situ loss rates indicated that uptake of both nuclides took place during restricted periods and that uptake and loss was relatively rapid. Use of a mathematical model relating specific activity of the organism to specific activity of the source at the time of shutdown indicated that sediments could not be ruled out as a radionuclide source to Anodonta. However, the low specific activity (5.8 nCi65Zn/g; 0.5 nCi54Mn/g) of a single organism transferred from the Willamette River to McNary Reservoir indicated that uptake from higher specific activity sediment was minor. Loss rates of 65Zn in several organs of Anodonta were in the order: fluids > muscle = shell > viscera > mantle > gills Interpretation of these data was complicated by the fact that uptake was probably taking place during the period of measurement.

* Major Professor:Robert L. Holton, Ph.D. 23

GROWTH RESPONSE OF A MARINE PHYTOPLANKTON COCCOLITHUS HUXLEYI, TO VARIOUS CHEMICAL FORMS OF COBALT* M.S. Thesis by Harold L. Longaker

The results of a preliminary experiment suggested thata complexed form of cobalt was more efficacious in promoting growth ofa marine phytoplankton than ionic cobalt.The phytoplankton used in this experimentwasCoccolithus huxleyi,a vitamin B12 producer, and the cobalt complex was cobalt (II)- ethylenediaminetetracetic acid [Co(II)-EDTA]. A review of the biochemistry of vitamin B12 indicates that a B12-producer might prefer, ifnot require, Co(III) instead of Co(II). Since someof the Co(II)-EDTA in the preliminary experiment might have become oxidized to Co(III)-EDTA, the observedstimula- tion of growth could havebeen due to Co(III)-EDTA,

Two experiments were performed to determineifCo(III)-EDTA is more efficacious in stimulating growth than Co(II)-EDTA0 Coccolithus huxleyi, grown in batch cultures with constant illumination,was used inboth experiments. One experiment had cobalt concentrations of 10 and 1 ug/1; the other had concentrations of 1 and 0.1 Vg/l. In both experiments there were no observed differences in specific growthrates between treatments of Co(III) as the EDTA complex with 10-6 M additional EDTA and Co(II) with 10-6 MEDTA. Both of these treatments resulted ina specific growth rate larger than controls without added EDTA or cobalt. It is not possible to measure the amount of Co(II)-EDTA that is oxidized to Co(III)-EDTAat the concentrations used in theseexperiments. Consequently these results cannot be used as a basis for rejecting the hypothesis that Co(III) is the requiredform of cobalt. Since Co(III)-EDTA without the additional 10-6M EDTA was apparently able to stimulate growth in relation to thecontrols, it is assumedthat C.huxZeyiis capable of utilizing this form of cobalt.

* Major Professor: Robert L. Holton, Ph.D. 25

MEETINGS ATTENDED*

NormanH. Cutshall

Gordon Conference on Chemical Oceanography, New London, New Hampshire, July 31-August 4, 1972.

*Marine Pollution Monitoring Workshop (NOAA), Santa Catalina, California, October 25-28, 1972.

Lead Workshop (IDOE), California Institute of Technology,Pasadena, September 16-20, 1973.

Heavy Metals in the Aquatic Environment, Nashville, Tennessee, December 3-6, 1973.

*PollutantTransfer Meeting(IDOE),PortAransas, Texas, January 11-12, 1974.

Gordon Research Conference on Chemical Oceanography, Santa Barbara, California, January 14-18, 1974.

Estuaries of the Pacific Northwest (4th Annual Technical Conference), Corvallis, Oregon, March 14-15, 1974.

Andrew G. Carey, Jr.

Lawrence Livermore Laboratory. Division of Radioecology. "Ecology and radioecology of benthic invertebrates of the Northeastern Pacific Ocean." Seminar presented March 5, 1973.

Western Beaufort Sea Ecological Cruise (WEBSEC) Conference, U.S, Coast Guard, Long Beach, California, April 25-27, 1973.

*Conference on Environmental Effects of Explosives and Explosions, Naval Ordnance Laboratory, Silver Springs, Maryland, May 30-31, 1973,

Workshop of U.S. Program for Oceanography in the Bering Sea (NSF Polar Programs), Salishan, Oregon, November 26-30, 1973.

*Symposium on Beaufort Sea Coastal and Shelf Research (The Arctic Institute of North America), San Francisco, January 7-9, 1974.

Planning Conference for the U.S. Program for Oceanography of the Bering Sea, Fairbanks, Alaska, February 21-24, 1974.

* Abstracts or texts of papers presented are included in the following section. 26

Robert L. Holton

Second International EstuarineConference, MyrtleBeach, S. Carolina, October 16-19, 1973.

Pollutant Effects Meeting(IDOE),CollegeStation,Texas. November 2-3, 1973. Estuaries of the Pacific Northwest (4th Annual Technical Conference), Corvallis,Oregon, March 14-15, 1974.

William G. Pearcy

AmericanSocietyof Limnology andOceanography,Salt LakeCity, Utah, June10-14, 1973.

TunaConference,LakeArrowhead, California, October 1-3, 1973. October 1-3, 1973.

*Eastern Pacific OceanicConference,LakeArrowhead,California, October 3-4, 1973.

*Abstracts of papers presented are includedin the followingsection. 27

PAPERS PRESENTED

BENTHICECOLOGY OF THEWESTERN BEAUFORT SEA CONTINENTAL MARGIN: PRELIMINARY RESULTS* A. G. Carey, Jr., R. E. Ruff, J. G. Castillo, and J. J. Dickinson

ABSTRACT The relationships between benthic organisms and the polar marine environment of the continental shelf and slope of thewestern Beaufort Sea are being defined by statisticalanalyses of faunal and environmental data. Of particular interest are the ecologicaleffects onbenthic community structure of the uniformly low bottom temperatures, the low and unpredictable input of food, and the scouring of the shallower continental shelf by ice.

Samples ofmeio- and macro-infauna and mega-epibenthos were collected during the 1971 and 1972Western Beaufort Sea Ecological Cruises (WEBSEC) on the USCGC GLACIER along a 333-km study area off the Alaskan North Slope betweenCape Halkett and Barter Island. Twenty-five otter trawlsamples, 20 bottomcameraruns, and 199 quantitativegrab samples were obtained at depths between 15 and 2600meters. Preliminary results based on data from twenty bottom trawl samples, 70 grabsamples,and bottom photographs demonstrate that species are restricted in their distribution within depthzones. The macro-infauna (5 1.0 mm) on the mid-continental shelf exhibita highly patchy distribution and variable taxonomiccomposition,implicating ice scour as a significant environmental .influence. The numerical density and biomass of the macro-infauna increase with depth and distance fromshore,reaching a maximum on the upper continental slope and decreasing to low values at depths of 2600 meters.

This benthicstudy,as a part of the description of the relatively pollution-free Beaufort Seaecosystem,presents a valuable baseline for future studies in the area.

Paper presentedat theSymposium on Beaufort Sea Coastaland Shelf Research(The Arctic Institute of North America),San Francisco, January 7-9, 1974. 28

BENTHIC ECOLOGICALSTUDIES AT DEEPWATERDUMPSITE G IN THE NORTHEASTPACIFIC OCEAN OFF THE COASTOF WASHINGTON* Andrew G. Carey, Jr., James R. Rucker, and Ronald G. Tipper

ABSTRACT As a part of the 1971 environmental survey of the Navy deep water munitions dumpsite G, several components of benthic fauna were sampled on a transect of five stations on the northern Cascadia Abyssal Plain in the Northeast Pacific ocean approximately 167 km west of the Strait of Juan de Fuca.From 1969 to 1970, five Liberty ship hulks, loaded with out-dated munitions, were scuttled in the dumpsite area.The major objective of the environmental survey was to determine if the sea floor environment and biota had been permanently damaged by the detonation during sinking of the five lots of munitions and by the presence of corrosive products from the debris and by-products from the explosions. Benthic organisms were sampled and studied as possible indicators of environmental quality.Infauna were collected by a 0.1 m2 Smith-McIntyre bottom grab, and mega-epifauna by a three-meter quantitative beam trawl. Certain mega-epifauna were studied by analysis of stereo bottom photographs taken along the trawl tracklines.Abundance and species composition were determined and were correlated with benthic environmental data including distance from the debris areas.There were no general, long-term effects of ordnance dumping at DWD-G that we could detect.There were indications that there may have been local environmental changes in trace metals immediately adjacent to the debris site, and perhaps some faunal changes within the debris area.There were no significant differences in faunal composition or abundance between the stations near the debriszone and at corresponding depths up to 20 nmi away. The Deepwater Dumpsite - G at the base of the Nitinat deep-sea fan on the Washington continental slope supports a more abundant mega-epifaunal community than is found in similar environments to the south, off Oregon.

*Paper presented at the Conference on Environmental Effects of Explosives and Explosions, Naval Ordnance Laboratory, Silver Springs, Maryland, May 30-31, 1973 29

MULTISPECTRALREMOTE SENSINGOF THE OCEAN OFF OREGON* William G. Pearcy andDonald F. Keene

ABSTRACT Multispectral scanner measurements from aircraft revealed complicated and often abrupt changes in ocean color off the Oregoncoast during the summer of1969. The changes in water color appeared less conservative than changes of sea surface temperature measured withan infrared radiometer, although the general areal patternswere similar. Color and thermal fronts, which often coincided, were most pronounced between upwelledwater, Columbia River plume water and oceanic waters.

Normalized spectra of backscattered light inseven bandwidths and differences between specific bandswere useful in identifying water types. The blue(.45-.47p)minus green(.52-.55u)differences were highest in oceanic water, intermediate in plume water and low in chlorophyll-rich upwelled water. The green(.52-.55p)minus orange(,58-.63p)differences demonstrated opposite trends and were highest inshore and lowestoffshore. Near the mouth of the Columbia River, where water containeda large load of particulate matter, both theblue-green and green-orange differences were low.

Occurrences of multiple color bands in the upwellingzone,and fronts along both inshore and offshore margins of the Columbia Riverplume,result from dynamic processes that may influence concentrationsof pelagic fishes.

*Presentedat theEastern Pacific oceanicConference,Lake Arrowhead, California, October 3-4, 1973. 30

DETERMINATION OF MERCURY IN MARINE ORGANISMS BY ATOMIC ABSORPTION SPECTROPHOTOMETRY *

J.R. Naidu and N.H. Cutshall School of Oceanography Oregon State University

This procedure is used for the measurement of the total mercurycon- tent of tissue samples from marine organisms. The analytical system on which the method has been developed is built arounda Perkin-Elmer Model 303 instrument fitted with a 20 cm absorption cell. The 300 cc volume of the closed-cycle gas phase affords a lower limit of detection of 1ng of mercury. For the procedure described this corresponds to 2 ng/g wet weight concentration in the sample.

This sample is digested with a combination of sulphuric and nitric acids and hydrogen peroxide in order to dissolve the tissue and to convert the mercury to inorganic Hg (II). An aliquot of the resultant solution is reacted with Sn (II) and hydroxylamine to reduce the mercury to Hg(0), and the mercury vapors are swept from the solution by an air stream. The air stream is cycled through a dessicant to remove water vapor and then through the absorption cell positioned in the optical path of a mercury hollow cathode lamp. The quantity of mercury present is determined by comparison of the resultant attentuation of the 253.7 nanometer line to the attenuation caused by standard mercury solutions. Digestion Step Apparatus: The digestion is carried out in a flat bottom flask, fitted to a reflux condenser (Friedrich). Heat is applied to the flask from below using a hot plate (Figure 4.13). Reagents: Concentrated Nitric Acid Concentrated Sulphuric Acid Hydrogen Peroxide 30% Approximately 5 to 10 grams of freshly thawed sampleare accurately weighed into the digestion flask.A few boiling beads are added, and the flask is attached to the reflux condenser. (If it is desired to report results on a dry or ash weight basis, another tissue subsample may also be weighed out for determination of dry:wet and ash:wet ratios.) Ten ml of concentrated nitric acid is added and allowed to stand overnight withno applied heat but with the condenser coolant water on. The temperature is then raised to 50° C for an hour. The flask is allowed to cool, and 10 ml of a 1:1 (v/v) mixture of concentrated nitric acid and concentrated sulphuric acid is added. The temperature is raised to 60°C for three hours. Ten

*Published in:Marine Pollution Monitoring Strategiesfor a National Program. Deliberations of a workshop heldat Santa Catalina Marine Biological Labora- tory of the University of Southern California,Allan Hancock Foundation, October 25-28, 1972. Sponsored by the National Oceanic and Atmospheric Administration of the U.S. Departmentof Commerce. October 1972.pp. 136- 143. 31

Hot plate

MERCURY DIGESTION APPARATUS (BIOLOGICAL) FIGURE 4.13 32

ml of 30% hydrogen peroxide is added in1 ml increments allowing the foam- ing to subside between additions. Temperature is increased to 80°C for 1 hour, after which the digestion iscomplete.

The flask is chilled withan ice bath and the condenser walls rinsed down with 50 ml of deionized distilledwater. The chilled solution is filtered through glass wool and dilutedto 100 ml in a volumetric flask using deionized distilled water. The digested solution may be kept under refrigeration (5-10°C) foras long as 6 months. The solution should be brought to room temperature just beforeanalysis.

Each batch of samples digested includesa reagent blank run as well as a standard run. The blank is 10 ml of deionized distilledwater, and the standard is from the workingmercury standard. No sample is added to either the blank or standardruns. Analysis Step Instrumentation: An atomic absorption spectrophotometer is fitted withan absorption cell as shown schematically in Figure4.14. A recorder readout is very convenient althoughnot ab- solutely necessary. The absorption cell should have a path length of approximately 20cm and a diameter of about 2 cm. Windows are quartz of not more than 0.3cm thickness. The water vapor trap is loosely packed magnesium perchlorate. The system is connected with 1 cm Tygon tubing. Reagents: 1 N Sulphuric Acid (Dilute28 ml of concentrated sulphuric acid to 1 liter.) Reducing Agent (Dissolve 100 ml ofconcentrated sulphuric acid, 5 g of sodium chloride, 5.3 g of hydroxylamine hydrochloride and 20 g of stannous chloride in sufficient deionized distilled water to makeone liter of solution. Reducing agent should be made fresh daily.) Mercury Stock Solution (Dissolve about 0.14g of mercuric chloride in 100 ml of 1 N sulphuric acid. The solution contains a nominal 1 mg/ml of Hg. Solution should be prepared monthly). Working Mercury Solution (Working standardsare prepared daily by successive dilution of the mercury stock solution in 1 N sulphuric acid to obtain a.nominal 100 ng/ml solution. This solution should be calibrated against a suitable standard. Five ml of the digested solution isadded to the reaction flask and sufficient 1 N sulphuric acid addedto bring the total volume to 20 ml. Three ml of the reductant solution is addedby pipette, and the flask is closed. The flask is gently swirled by handfor one minute and the air pump then started. If a recorder readout is used the absorbancewill rise to a maximum within 30 seconds and then falloff to a steady level after about one minute. When the absorbance reaches the steadylevel the bypass valve is opened and the system exhaustedthrough the mercury trap. When the absorbance reading has returnedto zero, the bypass is closed and a Absorbtion Cell

APPARATUS FOR THE ESTIMATION OF MERCURY FIGURE 4.14 34

volume of the working mercury solution is added to approximate the quan- tityof mercury found in the analysis run.If the quantity of mercury found is too high or too low for optimumprecision,the run is repeated using an appropriate volume of the digestedsolution. The reagent blank from the digestion step is generally carried through the analysis step with a batch of samples. Calculations The wet weight concentration of mercury in the sample is calculated as: (S-B)(Hg) refV C = R x W x 100 ml ,

where: S is the sampleabsorbance,B is the blankabsorbance,(Hg)ref is the quantity of mercury added in the reference solution, V is the volume of digested solution added, R is the reference absorbance and W is the weight of sample originally used. Note on Reagent Contamination Reagent grade chemicals are used through the total analytical scheme. Nevertheless,every batch of chemicals including those certified "mercury free" should be run forcontamination. The calculation given above makes no allowance for contaminated reducing agent.If the reducing agent gives an absorbance when run directly and cannot bepurified,this source of error should be taken into account in the calculations. Standards and Reference Materials Mercury standard solutions for calibration of the working standard are are available from Fisher Scientific Supply and from Beckman. Standard solutions containing inorganic and organic mercury are available from the Environmental Protection Agency.Since methyl mercury is much more toxic and has different environmental pathways than inorganic mercury, cases arise when one would like to know the fraction of the total mercury present as methyl mercury.

NON-CONSERVATION OF HEAVY METALS IN THE COLUMBIA RIVER ESTUARY*

NormanCutshall,RobertHolton,LawrenceC. Small

ABSTRACT

Manganese,copper and zinc in filtered waters are not conserved during mixing of Columbia River water withseawater. Metal concentrations at about 7%o salinity are higher than would result from conservative mixing. We suggest that the excess metals are derived from riverborne suspended matter. Reprecipitation of the metals seems to occur at about 20%o salinity.

Presented at the Pollutant Transfer Meeting(IDOE),PortAransas,Texas, January 11-12, 1974. 35

RESEARCH IN PROGRESS

POST-SHUTDOWN RADIONUCLIDE TRANSPORT STUDY

Vernon G. Johnson and Norman Cutshall

Sample collections for the transport study ofresidual Hanford radioactivity in ColumbiaRiver bed sediment were terminated in the lower River and estuary in September1972. Sediment sampling was terminated at McNary Reservoirin July 1973.

The primary effort since the last progressreport was devoted to gamma- ray counting and spectral analysis. Progress was severely hindered because of instrument drift and instability. Because of this, considerable effort was devoted to aquiring, modifying, and testing twocomputer programs that permit compensation for drift and gain shift. Tests with synthetic samples containing the major radionuclidespresent in Columbia River sediment showed good agreement between amounts addedand amounts reported by the computer program. The new method (developed at ORNL) also improvedprecision as well as accuracy. Using the present versionwe have been able to recognize increases of such nuclides as 54Mn in 12 cm3 sediment sampleswhere the new inputs are as low as 1 pCi/g. As of this writing, all samples have been counted and the spectral data reductionis now nearly complete. Some trace metal analyses remain before specificactivities in sediment can be calculated. Some salient features of the data thatare emerging are as follows:

1) Erosion of Bed Sediments and Associated Radionuclides

Hanford radionuclide concentrations in themonthly composite water samples for the lower Columbia (Prescott,Oregon) showed the characteristic spring maximum in both 1971 and 1972. The 1972 maximum, however, was substantially lower than can be accountedfor by decay alone. It seems likely that either the June 1971scour removed most of the mobile and available radionuclide inventory,or that new deposition of non-labelled sediment "shielded" the residual inventoryfrom further erosion/transport. McNary reservoir core data should aid inthese considerations.

2) Fallout from Atmospheric NuclearTests

In addition to the residual Hanford radioactivity,fallout from Chinese nuclear tests was clearly presentin many of the water samples. In the Prescott composite samples,two fallout concentration peaks occurred that seemed to be correlated more with thetiming of the spring freshet than with the actual dates the tests occurred. One explanation is that the fallout was stored in the high elevationsnow pack and was released in the spring melt. 36

Composition of the fallout also seemed to be somewhat unusual. For example, there was a marked enrichment of 103Ru over 106Ru and an apparent absence of140Ba_14OLain some rainwater samples collected shortly after the fall 1971 test. This is probably related to precursor/particle size/ atmospheric transport fractionation relationships. Lai and Freiling observed, for example, that the fine fraction of particulate debris downwind from the Kiwi reactor transient test was strongly enriched in 103Ru and depleted in the volatile fission products including 140Ba_14OLa.1 These authors noted that tThis was very similar to fission product behavior of bomb debris.

In addition, filtered river water samples consistently showed a high soluble 95Zr-95Nb fraction. As noted in the last report, this is also somewhat contrary to the expected behavior for this refractory fission product.

3) Possible Radionuclide Desorption in Columbia River Estuary

Although the data are fewthe filtered estuary water samples show significantly higher soluble6DZnthan do river water samples for similar collection dates. This observation tends to support Evans and Cutshall's2 earlier conclusion that seawater desorption of 6 Zn occurs from suspended material brought into the estuary from the river, and lends further support to the idea that the adsorbed state of 65Znis somehowdifferent for bottom sediment and suspended material. If this tentative conclusion is found to be firm, it may be cause for rejection of the hypothesis that there is interaction of exchange between the fine fraction of bed sedimentand suspended sediment, at least in the Columbia River System.

4) Evaluation of Europium Radioisotopes for Age-Dating

Europium isotope ratios (152/154) in McNary sediment were consistently much lower than would be expected or predicted from thermal neutron cross section data. Cross section data for the parent elements of these two isotopes show that the observed radionuclide ratio in sediments could be accounted for if the neutron spectrum during irradiation were rich in the epi-thermal region. Because this is a region of rapidly changing cross sections or resonances, this may place a serious limitation on the usefulness of the europium isotopes for age dating. More needs to be known about The neutron irradiation environment in which the europium isotopes were produced. In the absence of such knowledge, the assumption of constant "initial" ratios is grossly untrustworthy.

1 Lai, J. R. and E.C. Freiling. 1970. Pp. 337-351 in: Radionuclides in the Environment, American Chemical Society. 2 Reprint included in this report. 37

5) Observationof 125Sbin Sediments Antimony-125 was detected periodically in McNary sediments and was found to be non-homogeneously distributed in sediment samples.Its random occurrence, both with depthin cores and in areal extent in surface sediment, suggest it is present as "hot particles". It is also noteworthy that no other detectable associated radionuclides occur in those samples containing 125Sb inabnormal amounts orproportions. The apparent radiochemical purity suggests it may be an activation/corrosion product (rather than a fission product)involving a target material such as 124Sn and it was released to the river from some experiment or accident that contained only 125Sb.

6) Possible Chemical Release of 65Zn from Bed Sediment Zinc-65 in filtered water samples from below McNary Dam decreased rapidly following reactor shutdownand remainedlow throughthe summerand early Fall of 1971. However, samples collected in the winter months exhibited relatively high soluble fraction of65Zn. Either there was new input of Zn-65 or some chemical dissolution occurred in late fall andwinter. Uncertainty over possible new input from N-reactor makes this a difficult point to resolve. Earlier conclusions, based on specific activity data for soluble and particu- latephases,suggested that exchange equilibrium between 65Zn on suspended material and dissolved stable zinc was reached very slowly if at all.It may now be necessary to alter this conclusion.

EFFECTS OF SEAWATER ON METAL CHEMISTRY

NormanCutshall,DavidEvans*,Jerome J.Wagner,Norman Farrow

Mid-salinity estuarine waters contain higher concentrationsof certain metals (Mn,Zn, Cu) after filtration than do riveror seawatersfrom which they are presumably derived after allowanceis madefor linearmixing. We are seeking to determinethe causeof this non-linearityamong thefollowing possibilities: (1) mid-estuary inputs of metals,(2) desorption of metals from suspendedparticles,(3) variations in the colloidalsize material at the mid-estuary "turbidity maximum",(4) recyclingof metals in estuaries 'L la Odum's "nutrient trap". In addition to OSU studies in Oregon estuaries, Mr. Evans is carrying out his thesis research in a Southeasternestuary. He provides the following summary :

*Off campus1973,1974 school years. 38

Research is underway to develop a predictive model of dissolved concentrations of the trace metals Zn, Cd, Mn, Fe and Cu in estuaries of the southeastern United States. It is hoped these studies, based at the National Marine Fisheries Service, Atlantic Estuarine Fisheries Center, Beaufort, North Carolina, in collaboration with Ford A. Cross and Douglas A. Wolfe, can be compared with related efforts in estuaries of the northwestern United States by Norman Cutshall.

The model will use as input variables the physicochemical distribution of trace metals in adjacent ocean water and in river water flowing into the estuary. Deviations from the distribution predicted by a simple linear dilution model can then be related to in situ measurements of salinity, Eh and pH.

Progress to date has centered on development of a contamination-free field sampling method to separate dissolved and particulate trace metals. Efforts to separate trace metals from the interfering saline matrix of estuary waters have focused on the use of Chelex-100 exchange resin. This seems most capable of efficient, simultaneous extraction of the trace metals of interest in small (250 ml) samples with minimum contamination and relative simplicity. Flame- less atomic absorption spectrophotometry using a Perkin Elmer HGA2000 graphite furnace is the final analytical tool.

Other efforts have been expended in learning the particular characteristics of the Newport River estuary pre- requisite to successful sampling.

CYCLING AND DECLINE OF RADIOACTIVITY IN COLUMBIA RIVER ESTUARY AND IN ALDER SLOUGH BIOTA

William C.Renfro and NormanD.Farrow

Data collection for this study has been reduced to quarterly intervals and is expected to be phased out within one or two years.Dr. Renfro is presently Environmental Engineer with Northeast Utilities in Hartford, Connecticut.We are working with him toward publication of results. 39

VARIATIONS OF ZINC-65 IN ZOOPLANKTON RELATIVE TO THE COLUMBIA RIVER PLUME

Bruce E. Rieman andW. G. Pearcy

The radioactivity of macro-zooplankton (mainly euphausiids and copepods) collected during fourcruises,June-October1969,showed definite geographical and seasonaltrends. Zinc-65 radioactivity per gram and specificactivity was inversely related to distanceoffshore,except during the fall when the plume wasill-defined. Highest radioactivity near the coast was found during June and early July. In late July and August the plume was farther offshore and values of radioactivity and specific activity were relatively low. This trend for decreased accumulation of radioactivity with season continued into October when the lowest average valuesoccurred. Radioactivity was higher south of the mouth of the Columbia River than in the immediate zone of mixing.

CHANGES IN THERADIOACTIVITYAND THE SPECIFICACTIVITYOF ZINC-65 IN EPIPELAGIC,MESOPELAGIC AND ABYSSOBENTHIC ANIMALS OFF OREGON

W. G. Pearcy

Our extensive data on gamma emitters, and especially65Zn,in oceanic animals is being analyzed to describe specific activities associated with different species ofanimals. Most of these collections were made with non-closing nets; but with data from subsequent opening-closing net collections,we are able to assign approximate depths of capture to these animals. Changes in radioactivity with depth are of primary interest, especially rates of change of 65Zn specificactivities. These will be compared with that expected on the basis of physicalmixing,and differences in specific activity will be used to estimate biological vertical transport rates of 65Zn into the deep ocean. Preliminary comparisons reveal that the correlation coefficients between 65Znspecific activity and radioactivity of animals and their depth distributions arehigher for nighttime than daytime depth distributions. This suggests that depths occupied during the night are most important in the 65Zn uptake of migratory animals. This agreeswith Pearcyand Osterbergwho noted that 65Zn per gram washigher at night than day in the upper 150 m.1

1 Int. J. Oceanol. Limnol. 1(2): 103-116, 1967 40

CARBON-14 CONTENT OF DEEP-SEA ANIMALS

W. G. Pearcy and M. Stuiver

During our July 1973 cruise we collected 48 samples of animals and three samples of sediment for subsequent 14C analysis at the Department of Geological Sciences, University of Washington. Collections were made with closing midwater trawls to depths of 2400 m and with bottom trawls to depths of 3000 m. The objective of this study is to compare the age of carbon in deep-sea benthic animals with bathypelagic and mesopelagic animals. We hope this will provide insight into the relative vertical rates of carbon transport into the deep sea.

SPATIAL PATTERNS OF ZINC-65 SPECIFIC ACTIVITY IN PELAGIC AND BENTHIC FOOD WEBS ON OREGON'S CONTINENTAL SHELF, JULY 1970

H. A. Vanderploeg and W. G. Pearcy

During July 1970 extensive collections of pelagic organisms, benthic fishes and their stomach contents from a number of stations on Oregon's continental shelf were analyzed for 65Zn specific activity (SA). Also analyzed for 65Zn SA were water and phytoplankton from a station within the Columbia River plume.

Zinc-65 is conveyed to benthic fishes through detrital and pelagic food webs. Our analyses indicate that the relative importance of these pathways changes with distance from the Columbia River mouth. Furthermore, when these spatial patterns of SA in the food webs are coupled with 65Zn SA data for shelf sediments, water and phytoplankton, they will provide insight into the process responsible for the introduction of 65Zn into and its subsequent dispersal by benthic fishes and their food webs. 41

BIOLOGY AND SYSTEMATICS OF DEEP-SEAANIMALS OFF OREGON

D. Stein and W. G. Pearcy

Our study of the food habits of the macrouridfishes off Oregon revealed that these fishes are feeding generalists.1Of particular interest was the importance of prey commonly caughtin the upper 1000 m. This implies a direct food-link between mesopelagicand abysso-benthic animals that could be important in transport of elements,as well as energy, into the deep ocean.

Because it is important to have accurate identifications of animals that are radioanalyzed and because we are interested in the structure of deep-sea communities, some effortsare devoted to systematics of poorly studied taxa. At present David Stein is completing a study of the deep-sea liparid fishes. W. G. Pearcy, in collaboration with Dr. G. Voss of the University of Miami, is attempting to identify the deepwater cephalopods. The taxonomic literature on bothgroups is abysmal and at present the actual species of many of these animals inour collections are unknown.

DESCRIPTION AND BIOLOGY OF A NEW SPECIES OFBENTHEOGENNEMA (DECAPODA, ARISTAEINAE) FROM THE NORTHEASTERN PACIFIC

E. E. Krygierand R. Wasmer

The natantian decapod crustaceans comprisea large portion of the nektonicbiomassin theoceanic watersoff Oregon. The fourthmost abundant species of this groupis asyet undescribed and its biology is little known. The availability of seasonal collectionsand a large number of specimens had facilitated a biologicaland taxonomicstudy of thisspecies. Scrutiny of samples from other institutions has enabledus to determine the approximate geographical range of this species.It has a vertical distribution between 150 m to over 1000 m with diel changes in abundance above 400 m.Analysis of size classes had provided data on growth rate, spawning season and size at maturity of this new species.

1 Pearcy and Ambler, inpress; text included in this paper. 42

ISOTOPE DILUTION OF ZINC-65 IN OREGON COASTAL WATERS

NormanCutshall,IngvarL.Larsen

A model for 65Zn isotope dilution has been developed in which65Zn specific activity is represented in terms ofsalinities,stable zinc concentrations and source specificactivity. The model will be used to relate shoreline salinity data and specific activities in coastal organisms in order to search for alleged deviations from the "specific activity concept".

ZINC-65 AND IRON-55 TURNOVER IN COLUMBIA RIVER FISHES

NormanCutshall,C. David Jennings*

The loss of 65Zn and 55Fe specific activities in carp organs has been followed since shutdown of the Hanfordreactors. Different organs decline at different rates.We are attempting to compare direct analysis of loss data to turnover rates estimated from steady state specific activities.

LOSS OF ZINC-65 SPECIFICACTIVITYIN MUSSLES (MYTILUS CALIFORNIANUS)FROM THE OREGON COAST FOLLOWING REACTOR SHUTDOWN Ingvar L. Larsen and Jerome J. Wagner

Mussels from YaquinaHead,Oregon, some 175 km south of the Columbia River,were sampled periodically (bimonthly and latermonthly)over a three year period following the termination of the operations of the single-pass plutonium production reactors atHanford, Washington,during January 1971. The loss rate of 65Zn specific activity from a steady-state source and the time required for these coastal marine organisms to approach fallout background levels of 65Zn specific activity was determined. The existence of a long steady-state lag time observed following reactor shutdown indicated the presence of a large reservoir of 65Zn established within the Columbia River and marine environment.The decline was log-linear and uniform for over a two yearperiod,dropping from a steady-state value of 14.4 nCi 65Zn/g Zn during August 1971 to a value of 0.8 nCi 65Zn/g Zn as of November 1973.

* Oregon College ofEducation,Monmouth. 43

Examination of 65Zn specific activity data in mussels distributed along the Oregon Coast during the decline indicateda log-linear decrease in specific activity with distance south of themouth of the Columbia River. An apparent transport rate for 65Zn specific activitybetween Cape Kiwanda and Yaquina Head (58 km apart) during 23 July to 5 August 1971was also determined. Comparison of 65Zn specific activity between male and female specimens indicated that no discrimination basedupon gender existed. The data are currently being prepared for publication.

ZINC,RADIOACTIVE ZINC(65ZN),CADMIUM AND MERCURY IN THE PACIFIC HAKE,MERLUCCIUS PRODUCTUS(AYRES),OFF THE WESTCOAST OF U.S.

Janakiram R. Naidu and Norman Cutshall

The Pacific hake,MerZuccius productus(Ayres) was used to monitor the coastal waters off the West coast of the U.S. and Puget Sound for zinc, radioactive zinc (65Zn), cadmium and mercury.

This study has revealed that the Columbia River is not the source for zinc, cadmium and mercury, but was the main source of 65Zn as of January 29, 1971. Since then with the phasing out of all the plutonium production reactors at Hanford, the 65Zn activity in the river water, and therefore in the sea waters of the West Coast of the U.S., has approached minimum detection limits.

The 65Zn concentration in the hake reflect the position of the Columbia River water plume. With the decrease in sea water concentrations of 65Zn, as mentioned above, the activity of 65Zn in the Pacific hake has developed to levels beyond the limits of detection. Specific activity of 65Zn also follows a similar pattern and had declined with time (1969-1972).

Although Zn, Cd and Hg all belong to group IIB of the Periodic Table, no correlation was established between Zn and Hg, or between Cd and Hg. Zinc and cadmium concentrations were highly correlated.

Zinc and Cd concentration data increase with fish size approaching an asymptotic value at maturity. Mercury data exhibited a linear relationship with age and the slope was a function of location. The concentration factors follow the pattern Hg > Zn > Cd. Regulation is seen with reference to Zn and Cd, but in the case of Hg there is evidence that it is cumulative withage. 44

The utility of this study is seen in the establishment of the fact that the age distribution of fish caught commercially coincides with the maximum .attainable concentration of Zn and Cd. In the case of Hg, the accumulation is linear and continues to accumulate with age. In terms of environmental pollution hazards, this has important consequences based on public consumption of such fish with concentrations of Zn, Cd and Hg at their maximum possible levels in these fish.

LARGE BATCH CULTURE EXPERIMENTS IN RADIOISOTOPE KINETICS

Ingvar L. Larsen, Robert L. Holton, Norman Cutshall

We have begun a set of large batch phytoplankton experiments aimed toward determining the relative importance of factors such as chemical form and predation upon elemental and radionuclide cycling rates. Cross et al. found remarkably slow rates of assimilation of ionic 65Zn into established phytoplankton populations.1This slow assimilation could be due to failure of the ionic radioisotope to exchange with some "preferred" chemical form of zinc or to a very long population turnover time. We expect to test the former hypothesis by adding 65Zn to the medium before inoculation or fertilization and to test the latter by the addition of predators to the culture in order to accelerate turnover of cells.

NATURAL RADIONUCLIDE INDICATORS OF DENSE MINERAL DEPOSITS ALONG THE PACIFIC NORTHWEST COASTLINE

Ronald C. Scheidt, Norman Cutshall

Thorium and uranium content in sediments from the continental shelf off Oregon and Washington are being compared to dense mineral content and position relative to prehistoric lower sea levels. It appears that these nuclides are correlated with heavy mineral enrichments which occur in relict beaches from which light minerals have been winnowed out.

1Cross, Ford A., James N. Willis and John P. Baptist. 1971. Distribution of radioactive and stable zinc in an experimental marine ecosystem. J. Fish. Res. Bd. Canada 28: 1783-1788. 45

BENTHOS BENTHIC ECOLOGY AND RADIOECOLOGY

A. G. Carey, Jr.

The benthic ecological and radioecological research undertaken by our program has gone into its final phase of summary and analysis of data acquired in the eastern North Pacific Ocean off Oregon and Washington for the period June 1962 through August 1972.

The over-all research sponsored by the AEC has been aimed toward a further definition of the role of the sea floor in the oceanic ecosystem. Research has been particularly concerned with the deep-sea and near-shore environment under the influence of the Columbia River and other near- continental phenomena. Specific objectives have been: (1) to further define species composition, abundance and interrelationships of the benthic fauna;(2) to determine some of the sources, and rates of input of trace element, radionuclides and organic materials to thesea floor;(3) to further define the distribution and concentration of certain trace elements and radionuclides in bottom organisms in space and time, and (4) to determine the cycling of certain elements within the benthic environment, and between the benthic environment and other segments of the ecosystem.

Summary and Analysis of Benthic Radioecological Data

All data on radioisotopes and elements collected by our group from June 1962 through August 1972 are being summarized and statistically analyzed with the cooperation of the Oregon State University Department of Statistics and Computer Center. Over-all patterns of distribution and concentration of radionuclides in sublittoral, bathyal, and abyssal benthic invertebrate fauna are being described. Correlations are being made with depth, distance from the Columbia River, season, taxonomic levels, species, feedingtype, sediment type, periods of atmospheric atomic testing fall-out, and Hanford plutonium production reactor shutdowns to determine which features of the environment and the organisms influence the type and amount of radionuclide and trace element present.

Error and quality checks on the data now on file in the computer center are almost complete. Preliminary statistical models for determination of overall trends have been constructed and run. Over 3000 animal and sediment samples from Oregon and Washington waterswere collected and radioanalyzed by gamma-ray spectrophotometry. About a third of these include complementary stable element analysis for Zn by atomic absorption spectrometry for determination of 65Zn specific activities. 46

Summary of Cascadia Plain Benthic Ecology Research initiated under AEC support on the influence of continental proximity and of the Columbia River on the mega-epifaunal communities of Cascadia Abyssal Plain is beingcompleted. An initial introductory paper is under preparation and a Ph.D. thesis on the ecology of holothurians is inprogress. Of particular interest are the changes in faunal abundance and composition with increasingdistance,depth beingequal,from the con- tinentalslope. Changes in food source and quantity are hypothesized as thecause. Direct estimates of detrital organic particulate materials and their effect on abyssal benthic communities are being actively pursued on other projects. 47

RESEARCHCOMPLETED*

RLO-1750-62 Reprinted from J.Fish. Res. Bd. Canada 29(10):1419-1423 Sound-ScatteringLayersin the Northeastern Pacific

HENRY A. DONALDSON Graduate School of Oceanography University of Rhode Island, Kingston, R.I. 02881,USA AND WILLIAM G. PEARCY Department of Oceanography Oregon State University, Corvallis, Oreg., USA

DONALDSON, H. A., AND W. G. PEARCY. 1972. Sound-scattering layers in the northeastern Pacific.J. Fish. Res. Bd. Canada 29: 1419-1423.

Temporal and spatial variations in the depth, thickness,and vertical movement of scattering layers were studied usinga 38.5 kHz echosounder in waters off Oregon and between Hawaii and Adak, Alaska. Off Oregon one or two layers migrated towardthe surface at dusk and descended into deeper water at dawn. Although individualscattering layers migrated at a fairly constant rate during a twilight period, rates of ascent and descentvaried from day to day. Average rates were 2-3 m/min. During the night the depth of the top ofthe scattering layer averaged 40 m, and it varied between 50-245 m during the day.The bottom of the deepest recorded layer was 450 m deep. Regardless of intensity, the thickness of thescattering layers in the water column was usually greater during night than day. Itwas greater over the continental slope than abyssal depth, and in some years it was greater duringthe summer than during other seasons. Between Hawaii and Adak, Alaska, centralwaters had large amounts of scattering but low midwater trawl catches, transitionalwaters had high-intensity scattering layers and high biomass, and subarctic waters hadvery reduced scattering layers but high biomass.

DONALDSON, H. A., AND W. G. PEARCY.1972. Sound-scattering layers in the northeastern Pacific.J. Fish. Res. Bd. Canada 29: 1419-1423.

Nous avons etudie les variations temporelleset spatiales de la profondeur, de 1'epaisseur et des deplacements verticaux des couches diffusantesan large de l'Oregon et entre Hawaii et Adak, Alaska. Nous avons utilise a cette finun sondeur a echo de 38.5 kHz. Au large de l'Oregon, une ou deux couches diffusantes remontent vers ]a surfacean crepuscule et retournent en profondeur a l'aube.Quoique les couches individuelles se de- placent a une vitesse relativement constantependant une periode de demi jour, le rythme de montee et de descente varie de jouren jour. La vitesse moyenne est 2-3 m/min. La profondeur de la partie superieure de la couchediffusante est de 40 m en moyenne la nuit, et elle varie de 50-245 m le jour. Le niveau inferieurde la couche la plus profonde a ete enregistre a 450 m. Independamment de l'intensite, lescouches diffusantes dans la colonne d'eau sont ordinairement plus epaisses la nuitque le jour. Elles sont plus epaisses sur la plate-forme continentale que dans la zone abyssaleet, certaines annees, plus epaisses en ete qu'en toute autre raison. Lors d'une croisiere entre Hawaii et Adak,Alaska, nous avons observe un haut degre de diffusion, mais de faibles captures du chalutsemi-pelagique dans les eaux centrales, des couches diffusantes de haute intensity etune biomasse abondante dans les eaux de transition, et des couches tres reduites, mais a biomasse abondante,dans les eaux subarctiques. Received October 7, 1971

SINCE World War II, sound-scattering layers have Organisms suggestedascauses of scattering been discovered in midwater regionsof most of the include squids (Lyman 1948), euphausiids (Moore world's oceans. Many studies have beenaimed at 1950), fishes (Marshall 1951; Johnson et al. 1956; determining the cause of scattering,but few haveBackus and Barnes 1957; and Taylor 1968), and examined seasonal and spatial variationsin one siphonophores (Barham region. 1963).Tucker(1951), Kampa and Boden (1954), and Barham (MS 1956, 1966) suggested that scattering may be caused by Printed in Canada (J2321) more than one group of organisms. The few seasonal * Arranged by AEC document number. 48

1420 JOURNALFISHERIES RESEARCHBOARD OF CANADA, VOL. 29, NO. 10, 1972

studies of scattering layers are those by Barham DAY-NIGHT SCATTERING (MS 1956) in Monterey Bay, California, and Hun- kins (1965) from Fletcher's Ice Island. Several series of echosoundings were made at one layers and the station for periods of 10-48 hr to observe differences The migrations of scattering in the vertical position of the layers and changes in changes in scattering from day to night, from season to season, and from inshore to offshore were studiedthe thickness of scattering in the water column. off the Oregon coast over a period of several years. As expected, the main scattering layers were The relations of scattering to the biomass of various deeper during the day than the night (Fig. I A and B). groups of animals were investigated on a cruise Excluding surface scattering, the average depths and standard deviations for upper layers (measured from between Hawaii and Adak, Alaska. the surface to the upper edge of scattering) were 40 f 14 m at night and 180 f 52 m during the day. Materials and Methods The maximum depth of a recorded scattering layer (measured to the lower edge of scattering) was 450 Most of this study was made at stations 25 and 50 nautical miles off Newport, Oregon. The 25-mile station M. is located over the outer continental shelf, and the 50- Generally, the total thickness of scattering layers, mile station is at the outer edge of the continental slope regardless of intensity, was greater during night in oceanic waters. Observations were also made on a (140 d= 57 m) than day (109 ± 66 m) (Fig. 1B). In cruise between Hawaii and Adak, Alaska. other echograms, day and night intensities and To detect scattering layers a model 510-5 Simrad amounts of scattering were similar (Fig. IA). Echosounder, operated at a frequency of 38.5 kHz, was The observed greater thickness of scattering layers used. From January 1963 to September 1964, echotraces at night could be due to their proximity to the were recorded aboard the R/VAconausing a pulse length echosounder or to the spreading of the layers at night of 2.4 m/sec, an amplification of about 54 db, and ain the absence of light intensities to which the or- gain reduction in the surface waters. After September, ganisms are adapted during the day. The spreading 1964, echotraces were taken aboard the R/VYaquina where the setting on the echosounder was the sameof the layers at night is supported by Taylor (1968), except for a slightly lower amplification (about 6 db who caught fewer fishes in scattering layers during lower). the night than during the day. To examine possible relations of scattering to the biomass of various groups of organisms between Hawaii and Adak, Alaska, animals were collected with a 6-ft INSHORE-OFFSHORE VARIATIONS IN SCATTERING Isaacs-Kidd midwater trawl (with a 5-mm square measure Intense scattering was often present during the mesh liner). The trawl did not sample specific layers butnight from the surface to the bottom over the con- was towed obliquely from 0-200 m at night. The echo- tinental shelf (Fig. IC, station 15 miles from shore). sounder was operated continuously while sampling was Near submerged banks the midwater scattering layer underway. was often penetrated by the bank while inother areas dense clouds of scattering (possibly rockfishes, Results Sebastodes spp.) on the top of the banks corresponded with a gap in the overlying, midwater VERTICAL MIGRATION OF SCATTERING LAYERS scattering layer (Fig. 1E). Isaacs and Schwartzlose The vertical migration of scattering layers off(1965) suggested that a similar gap was caused by Oregon consisted of the typical descent into deep the feeding of predators from the bank on the midwater at dawn and ascent at dusk. During ascent water scattering layer. and descent one or two layers were observed to In waters beyond the slope the maximum thickness migrate. of scattering in the water column was often found The average rates and standard deviations are within 45 miles of shore and decreased in an offshore 2.4 ± 0.8 m/min for descent and 3.1 f 0.8 m/min direction (Fig. 1D). A maximum in the biomass of for ascent. The rates at which single layers migrated fishes, squids, and shrimps 45 miles offshore (Pearcy generally remained fairly constant. However, in unpublished) suggests a general correlation between some instances when two layers ascended, the lower biomass of small nekton and scattering off Oregon. sometimes migrated at an increasing rate. In one case the ascent of the lower layer slowed as the night SEASONAL VARIATIONS IN SCATTERING depth was reached. Although Bary (1967) distinguished four stages of migration based on changes Seasonal variations in the thickness and intensity in the rate of movement and the appearance of layers of scattering layers were examined in night echo- during migration, similar stages were usually not grams at the station 50 miles offshore. In 1963, apparent in our study. the amount of scattering was lowest in the winter 49

arrro" 2JEW-j a X. aui,nr a W

540-s15 t1:3;2sI ygs MILES FROM SHORE

BOTTOM

E FIG. 1.(A-B) Day versus night differences in scattering 50 milesfrom New- port,Oreg. (C-D) Inshoreto offshore variations in scattering off Newport, Oreg.(E) Scattering layers near a submerged bank off Newport, Oreg. Donaldson and Pearcy - J. Fish. Res. Bd.Canada so

APR - AUG OCT DEC FED MAR APR JUN JUL n=-r _

FIG. 2.(A-C) Seasonal variationsin nighttime scattering 50 milesfrom Newport, Oreg. in 1963 (A), 1964(B), and 1965 (C).

500 FIG. 3.Geographic variations in scattering between Hawaii and Adak, Alaska.

Donaldson and Pearcy - ].Fish. Res. Bd. Canada 0 z TABLE 1.Relationship of scattering to biomass of various groups between Hawaiiand Adak, Alaska(Samplingdevice: 6-ft midwater trawl). 0 Ratioof z % of the water euphausiids to column sampled Plankton total the other Shrimp by the trawl (g wet wt/103 m3 abundant Fish biomass biomass Squid Trace No. g dry biomass g dry in and occupied bytaken in the 6-ft plankters g dry wt/103 m3 wt/103 m3 FIG. 3. Date Position scattering trawl indicated wt/103 m3

4:1 Jelliest 0.43 0.11 0.05 1 June1, 1966 25°46.2'N 50 4.23 160°17.3'W 1:4 Copepods 2.56 0.70 0.64 2 June 12, 1966 33°33'N 25 28.70 164°21.6'W 16.63 3.08 1.85 3 June 16, 1966 42°35.5'N 75 35.55 2:3 Jelliest 171°02.8'W 0.50 4 June 22, 1966 50°29.0'N 0.5 64.40 4:1 Jellies' 3.32 1.15 176°04.O'W

'Jellies includes: salps, ctenophores, and medusae.

.p 52

1422 JOURNALFISHERIES RESEARCH BOARDOF CANADA, VOL. 29, NO. 10, 1972

and spring and highest in the summer and fallarctic water mass had almost no scattering but high (Fig. 2A). The total thickness of scattering wasbiomass of most groups including euphausiids. directly related to the number of reverberation layers. These results indicate the difficulty in correlating Most scattering occurred in the summer when fourscattering and biomass where different ocean en- reverberation layers were present. In 1964, scatteringvironments are sampled. Undoubtedly, more con- was quite variable. Moderately thick layers weresistent results would be obtained with discrete depth present during spring, summer, and winter (Fig. 2B).sampling and an analysis of species and their scatter- During July, the layers were not as thick but scatter-ing properties. ing intensity was high. These narrow, high-intensity layers may indicate a density of scattering agents Discussion equal to that of broad low-intensity layers, thus making evaluation of seasonal variations difficult. Variability was the most striking feature of the The data for 1965 were limited, but again indicated scattering off Oregon. The extent of migration of less scattering during the spring than during the the layers, the thickness of scattering layers present summer (Fig. 2C). over a diel period, and the thickness of scattering present during different months all fluctuated. This MID-PACIFIC OBSERVATIONS variability may have several causes: differences in acoustical scattering associated with differences in On the cruise from Hawaii to Adak, Alaska,species composition; changes in distribution, be- differences in scattering were related to water mass havior, and morphology of life stages of organisms; types and biomass. The scattering layers typical patchiness in the distributions of organisms; differ- of North Pacific central water are shown in Figureential attenuation of sound; and variable reactions 3, Traces 1 and 2. In the southern portion of thisof scattering organisms to physical conditions such water mass, well developed reverberation layersas ambient light. were present (Fig. 3, Trace 1); but, farther north, Although no marked seasonal changes in species scattering was less evident and occurred only as a composition of midwater fishes were observed off single layer (Fig. 3, Trace 2). Also, in the southern Oregon (Pearcy 1964), seasonal changes in number portion of the water mass, plankton biomass wasor biomass occur and may lead to seasonal differ- low and euphausiids were the dominant animals ences in scattering. Pearcy and Laurs (1966) caught (Table 1, Trace 1). Towards the northern half of thelarger numbers of mesopelagic fishes during June- water mass, plankton biomass increased notably October than during the remainder of the year at the and the dominant organisms changed from eu-station 50 miles from shore, the same station where phausiids to copepods (Table1, Trace 2). Thewe studied seasonal variations of scattering. The decrease in scattering is associated with this change biomass of small nektonic animals (fishes, squids, in dominant organism. and shrimps) was also larger during the summer The transition region between central and sub-than the winter over the continental slope (Pearcy arctic water masses was characterized by an increaseunpublished). In another study, Hebard (MS 1966) in scattering (Fig. 3, Trace 3) and biomass of allfound larger populations of Euphausia pacifica groups (Table 1, Trace 3). Biomass of shrimps, fishes,through the summer and fall of 1963 than in the and squids were largest here, and plankton biomass winter of 1964 at a station 45 miles from shore. was much larger than in central water. Patchy distribution of organisms may influence North of the transition zone, the Pacific subarctic scattering at one location, particularly if observations water mass was characterized by the virtual absence are several hours apart. At the same location where of scattering (Fig. 3, Trace 4). Nevertheless, catchesthe scattering layers were monitored, Pearcy (1964) contained the highest plankton biomass, composed found clumped distributions of four species of meso- primarily of euphausiids (Table 1, Trace 4), a lowerpelagic fishes. Patchiness was not apparent from biomass of fishes than that found in the transitionecho-traces, however. Layers appeared continuous region but higher than that in central water, and a when the vessel steamed between stations or was biomass of shrimps and squids comparable to thatadrift, suggesting that the volume of water from found in central water. which scattering was dectected was greater than the In summary, the central water mass had a largepatch size of animals. percent of the water column sampled by the trawl occupied by scattering but a low biomass of plankton, fishes, shrimps, and squids. Euphausiids pre- Acknowledgments dominated where scattering was at a maximum. This research was supported by the NSF (GB-1588 The transitional region had a large biomass of alland GB-5494) and the AEC (AT(45-1)1750, RLO- groups and a large amount of scattering. The sub- 1750-62). 53

DONALDSON AND PEARCY: SOUND SCATTERING LAYERS 1423 BACKUS, R. H., AND H. BARNES. 1957.Television- camera studies of mid-water sound scatterers.Deep- echosounder observations of midwater soundscatter- Sea Res. 3: 266-272. ers.Deep-Sea Res. 4: 116-119. BARHAM, E. G. MS 1956.The ecology of sonic KAMPA, E. M., AND B. P. BODEN.1954.Submarine scattering layers in the Monterey Bay illumination and the twilight movement of a sonic area, California. scattering layer.Nature 174: 869-871. Ph.D. Thesis. Stanford Univ., Stanford, Calif. 182p. 1963.Siphonophores and the deep scattering layer. LYMAN, J.1948.The sea's phantom bottom.Sci. Science 140: 826-828. Mon. 66:87-88. 1966.Deep scattering layer migration andcom- MARSHALL, N. B.1951.Bathypelagic fishes as sound position: observations from a divingsaucer.Science scatterers in the ocean. J. Mar. Res. 10: 1-17. 151: 1399-1402. MOORE, H. B.1950.The relation between the Eu- BARY, B. M.1967.Diel vertical migrations of under- phausiacea and the scattering layer.Biol. Bull. 99: water scattering, mostly in Saanich Inlet, British Co- 181-212. lumbia.Deep-Sea Res. 14: 35-50. HEBARD, J. F. MS 1966. PEARCY, W. G. 1964.Some distributional features of Distribution of Euphausica mesopelagic fishes off Oregon.J.Mar. Res. 22: and Copepoda off Oregon in relation to oceaniccon- 83-102. ditions.Ph.D. Thesis. Oregon State Univ., Corvallis, Oreg.85 p. PEARCY, W. G., AND R. M. LAURS.1966.Vertical HUNKINS, J.1965.The seasonal variation in the sound- migration and distribution of mesopelagic fishes off scattering layer observed at Fletchers Ice Island (T-3) Oregon.Deep-Sea Res. 13: 153-165. with a 12-kc/s echo sounder.Deep-Sea Res. 12: TAYLOR, F. H. C.1968.The relationship of midwater 879-881. trawl catches to sound-scattering layers off the coast ISAACS, J. D., AND R. A. SCHWARTZLOSR.1965.Mi- of northern British Columbia.J.Fish. Res. Bd. grant sound scatterers: interaction with thesea floor. Canada 25: 457-472. Science 150: 18/0-/813. TUCKER, G. H. 1951.Relation of fishes and other JOHNSON, H. R., R. H. BACKUS, J. B. HERSEY, AND D. organisms to the scattering of underwater sound.J. M. OWEN.1956.Suspended echo sounder and Mar. Res. 10: 215-238. 54 RLO-1750-84 Reprinted fromTrans. liner. Fish. Soc.102(2):317-322 Radioactivityin JuvenileColumbiaRiver Salmon: A Model toDistinguish Differences in Movement and Feeding Habits

GERALD P. ROMBERG' AND WILLIAM C. RENFRO2 Department of Oceanography, Oregon State University Corvallis, Oregon 97331

ABSTRACT The 32P specific activities and 65Zn concentrations were measured in juvenile salmon collected from the Columbia River during winter and spring of 1969. An environmental uptake study involving marked juvenile chinook salmon indicated that 32P specific activity values reach a maximum sooner than values for 65Zn concentration. The uptake study also suggested a short lag period before hatchery fish released to the Columbia River began rapid uptake of 322P and "'Zn. A graphical model based on 32P specific activities and"Zu concentrations was constructed which allows fish to be classified and differentiated with regard to their past feeding or movement behavior.Use of the model indicated that some hatchery fish released to the river either had feeding behavior unlike other hatchery fish or were not in the Columbia River the entire time from release until collection. Application of the model to a sample of 24 un- marked juvenile salmon collected from the Columbia River revealed several different groups of fish.

INTRODUCTION This report suggests a means to learn more In recent years hatchery rearing of salmon about one important phase of the salmon life has become an important factor in the sportcycle, the movement and feeding behavior of and commercialfisheriesofthePacific young hatchery-reared fish as they migrate Northwest. Great numbers of juvenile salmon to sea. are released from hatcheries to supplement Radioisotopes of several biologically sig- the native stocks. An important factor bear-nificant elements were introduced in small ing on the survival of hatchery-reared salmon concentrations into the Columbia River dur- is the time and size at which they are released ing the operation of plutonium production (Wallis, 1968). The subsequent history and reactors at Hanford, Washington. The ma- behavior of the young fish in their down- jority of these radionuclides were produced by stream migration to the ocean may determine neutron activation of mineral elements in the the success of the hatchery program. coolant water and consist of both gamma- and In the Columbia River Basin, salmon frombeta-emittingradionuclides.Phosphorus-32 a number of hatcheries supplement nativewith a half-life of 14.2 days and zinc-65 with chinook(Oncorhynchus tshatwytscha)and a half-life of 245 days were two of the biologi- coho(0. kisutch)salmon runs which havecally important radionuclides present in Han- been adversely affected for various reasons.ford reactor effluent.Phosphorus-32 decays The value of the hatchery program is mea-by beta emission with no gamma emission, sured by the numbers of hatchery-rearedwhile zinc-65 decay is 49% EC with a 1.12 adult fish caught by sport and commercialMeV gamma, 49% EC directly to the ground fishermen plus those which complete their state,and 1.8% positronemission.Both life cycle by returning to their natal stream.nuclides were concentrated by most Columbia To improve the operation of the hatcheryRiver organisms(Fosterand McConnon, program, it is essential to understand all parts 1962). of the life cycle of the artificially-reared fish. An unusual opportunity to observe the uptake of 32P and 65Zn by juvenile chinook Present address:Argonne National Laboratory, salmon in the natural environment occurred Argonne, Ill nois 60439. when 200,000 marked fish were released to 2 Presentaddress:InternationalLaboratoryof MarineRadioactivity, Musee Oceanographique, Prin- a tributary of the Columbia River from a cipalityof Monaco. hatcheryatRingold, Washington,located 317 55

318 TRANS. AMER. FISH. SOC., 1973, NO. 2

Willamette River * Indicates sampling site 0 24 48 ScaleinKilometers

FIGURE 1.-Map of lower Columbia River showing the locations of dams and fish sampling sites. about 25 miles downstream from the Hanford formalin solution.Prior to radioassay, the reactors. Samples of these fish were collectedalimentary tract contents were removed and downstream from Ringold over a period ofthe fish were vacuum dried at 60 C for 24 3 weeks and the 65Zn concentration (65Zn hours.For purposes of radioactivity decay activity/g dry wt) and 32P specific activitycorrections, the fish were assumed to have (82P activity/g total phosphorus) of each fish entered the Columbia River at 2400 hours, 12 were determined. On a plot of 32P specificMay 1969, and to have been captured at 1200 activity vs. 65Zn concentration, a graphical hours on the date of collection. set of boundaries (or model) was constructed On 23 March 1969, a collection of 24 un- to allow fish with differences in either foodmarked salmon ranging from 8.3 to 13.4 cm source or feeding behavior to be identified (standard length) was taken with a small based on their 65Zn concentration and 32Pseine from the Columbia River near Board- specific activity. man, Oregon. They were packed in ice and returned to the laboratory, where muscle-bone MATERIALS AND METHODS samples were dried at 100 C for 24 hours and At the Ringold hatchery, 200,000 age-group radioassayed. 0 chinook salmon were cold branded on the All fish were radioassayed for 6Zn indi- anterior left side by a small super cooledvidually by our placing the dried specimen branding wire. At 2100 hours on 12 Mayanterior end first into a 12-cc plastic tube 1969, these fish were released into a tributary and gamma counting for 100 minutes in a stream approximately 300 m from the Co- 12.7 X 12.7 cm NaI(Tl) well crystal coupled lumbia River (Fig. 1). A sample of these toa 512 channel analyzer.After gamma hatchery-reared fish was taken at the time ofcounting was completed, each fish was re- release and in the ensuing 3 weeks other sam- moved from the counting tube and dissolved ples of these marked fish were collected at in nitric acid for chemical extraction and con- two downstream dams by biologists of the centration of phosphorus.Preliminary pre- U. S. National Marine Fisheries Service. Thecipitationas ammonium phosphomolybdate fish, ranging from 6.8 to 10.3 cm standard was followed by a purification precipitation length, were dipnetted from the gate wells ofas ammonium phosphate.Final precipitates the dams and immediately preserved in 10%were collected by filtration, transferred to 2.5 56

ROMBERG AND RENFRO-RADIOACTIVITY IN SALMON 319

cm X 0.8 cm stainless steel planchets,dried, .91.110.11.11.4 M0.11.. O.w . 91* 0211.<..I 0! M.N.1$ D.n weighed, and counted for 32P with a low . Fi.11 00.....4 .r R.1n try background beta counter in conjunction with a time interval printer. A portion ofthe dried precipitate was dissolved in HCl and analyzed for total phosphorus by the common molybdenum blue colorimetric technique to determine percent phosphorus in the precipitate. Both 65Zn and 32P activities were corrected for decay to the time of collection. o d s 1x 1 w Ie zo zx z Phosphorus-32 levels in fish were expressed DAR AFTER RELEASEI FROM RINOOLD -32P/gram ixo as 32P specific activity (nanocuries 91,2 0./1./121. .. 0.111. O.w phosphorus)because specificactivityisa ..,.. 1.1,.11...1..11 . , sensitive measure of changes in the radio phosphorus content within a fish, and total stable phosphorus could be determined fol- eo lowing 32P analysis.Zinc-65 levels were ex- pressedas65Znconcentration(picocuries 65Zn/gram dry weight)because onlythe radioactive 65Zn could be determined by non- destructive gamma analysis prior to process- $ 10 12 14 1! 1! x0 xx x. ing for 32P content. DAYS AFTER RELEASE FROM RINOOLD

FIGURE 2.-Radioactivityof markedjuvenile RESULTS salmonvs. days spent in the Columbia River. Solid curves show trends for fish collected at The Dalles Uptake Study Dam while dashed curves are for fish collected at The increase of 32P specific activity and McNary Dam. 65Zn concentration in recaptured hatchery fish is plotted in Figure 2.Uptake curvesDam, far down river from McNary Dam, have been drawn through the median valuemust have passed through McNary Dam for each sample group. Fish contained littleabout 3 days earlier, when marked fish first 32Por 65Zn before release with -measured appeared and were collected at McNary Dam. values of less than 0.2 nanocuries 32P perComparison of the 32P specific activities and gram phosphorus (nCi 32P/g P)and less65Zn concentration in the first McNary Dam than 0.2 picocuries 65Zn per gram dry weightsample with the uptake curve for The Dalles (pCi 65Zn/g dry wt).Rise in 32P specificDam samples indicates a time lag before fish activity in The Dalles Dam samples (solidstarted rapidly accumulating 32P and 65Zn. curve in Fig. 2) was rapid and appeared toThis time lag could reflect the time required be approaching an equilibrium value.In- for a hatchery fish to adjust to the river con- crease of 65Zn concentration inThe Dallesditions and to begin feeding on natural food. Dam samples was also rapid, but failed to However, low 32P specific activities and 65Zn approach equilibrium during the samplingconcentrations in the second set of McNary period.Fish samples collected at McNarysamples (collected 15.5 days after release), Dam (dashed curve in Fig. 2) had a muchsuggests that additional factors other than a lower rate of increase in both 32P specificshort acclimation period may be responsible activity and f 5Zn concentration than for sam-for the lower radioactivity levels. ples collected at The Dalles Dam. Uptake of 32P and 65Zn in fish is mainly The first set of samples collected at eachthrough assimilation from their food (Watson d am was taken the first day a large numberet al., 1969). Large differences in the levels of Ringold marked fish appeared at the dam.of 32P and 65Zn in fish, therefore, must be Marked fish that first appeared at The Dallesprimarily due to differences in the quantity 57

320 TRANS. AMER. FISH. SOC., 1973, NO. 2

z activityinfoodbothaboveandbelow r McNary Damisalsounknown.Reduced ( EC O..ei. levels of32P and 65Zn would bepresent in aquatic foodorganisms that were either in tributariesalong the Columbia River or had ..fY.ainy A,Yp only recently entered the Columbia River. The Snake River, which contributes about n.. Rp..nm. .. CeWmai.Ri ... from a less Podia active Environment one-thirdas muchwater to the upper end of McNary reservoirasdoestheColumbia River, is the largest tributary to influence 65Zn CONCENTRATION (RCi Dz,,5 DRY WY.1 Ringold hatchery fish upstream from McNary DAYS AFTER RELEASE LOCATION Dam (U. S. Geological Survey, 1968). Also, ].S Ns Mary ..S 0.11.. the SnakeRiver plume, extending along the 1s., 0a11.. 04 NSNery eastern bankof upper McNary reservoir, I..S 0.11.. constitutes an environmentin which fish and food organismswould be exposed to lower levels of radionuclidesthan in undiluted Co- lumbia Riverwater. Furtherinvestigation is requiredto determine whether feeding behavior or influenceof a less radioactive environment isresponsible for the differences in the 0 20 40 60 00 100. 120 1.0 180 IW increaseof 32P specific activity and 65Zn con- 65Zn CONCENTRATION (1a

ROMBERG AND RENFRO-RADIOACTIVITY IN SALMON 321

the equilibrium level for G'Zn concentration. Within a fish population, which has had a similar past history, there will be some natural range of variability in the levels of32P and 65Zn accumulated by individuals. In the graphical model, asetof boundaries are drawn to encompass this natural range of variability. A point outside the arbitrary graphical boundaries with a 32P specific activity low . 40 10 60 70 60 f0 100 in relation to the65Zn concentration might 69ZnCONCENTRATION IpCi65z-/0 ORY WT-1 occur if a fish stopped feeding or was re- FIGURE 4.-Application of the model to distinguish stricted in its food consumption, because its several different groups of juvenile salmon collected 32P specificactivity would decrease more near Boardman, Oregon. Solid lines representbound- rapidly, than its s&Zn concentration (due to aries of variability for hatchery fish.Dashed curses are boundaries which appear to apply to Boardman the shorter physical half-life and biological fish. turnover of 32P) .The same results would occur if a fish moved from the Columbia River into less radioactive waters for a timeand °°Zn. A food source low in 32 P and 65Zn and then returned to the river. There appearscould be obtained in the hatchery stream or to be no way to distinguish between the twofrom a less radioactive tributary such as the factors; therefore, a point outside the graphi-Snake River. Points representing fish number cal boundaries is attributed to either atypicalfive (and possibly number four)fall below feeding behavior or return to the Columbiathe boundary set for variability, suggesting River from a less radioactive environment.that these fish either had feeding behavior unlike the others or returned to the Columbia Application of the Graphical Model from a less radioactive environment. Radioactivity levels present in muscle-bone Increase of radioactivity in recaptured samples of 24 juvenile salmon collected near hatchery fish follows the expected patternBoardman, Oregon, are shown in their rela- with each set of samples grouped togethertion to the model in Figure 4. The dashed chronologically, except for the second set oflines are arbitrary boundaries which appear samples from McNary Dam (numbered oneto apply to this population. There are at least through five) which have levels lower thantwo reasons why the boundaries for Board- those in fish collected earlier at The Dallesman fish should differ from thoseestablished Dam. Boundaries for natural variability in forrecapturedhatcheryfish. First,the Figure 3, therefore, were set by consideringBoardman sample was collectedin March all values except those of the second McNarywhen the levels of radioactivity in fish are Dam sample. A representative standard devi-usually low due to decreased food consump- ation due to radioactivity counting error istion and reduced radioactivity in the food shown for one point plotted in Figure 3 (The(Davis and Foster, 1958; Foster and Mc- Dalles Dam, 15.5 days after release). Connon, 1962).Secondly, preparation for Points for McNary Dam fish numbers oneradioassay was different and the values are through three fall within the set boundaries.for muscle and bone only, not the entire fish. However, their location in the model relative Although the boundaries must be arbitrar- to the location of the first Dalles Dam samples (designated by X's) indicates that theily set by inspection, they serve as a guide levels of 32P and °tZn accumulated are equiv- to distinguish groups of fish which appar- alent to a residence time in the Columbiaently have had differentrecenthistories. River of only 6 to 10 days. This implies thatThere is a group of nine Boardman fish (Nos. the fish either had reduced food consumption 1, 2, 3, 4, 6, 7, 13, 15, and 17) which were the first week or consumed food lower in 32Prelatively new to the Columbia River with 59

322 TRANS. AMER. FISH. SOC., 1973, NO. 2 both low 32P specific activities(below 1.0possible to establish boundaries. Use of the nCi/g P) and low 6SZn concentrations (belowmodel not only helps distinguish between 12 pCi/g dry wt). Radioactivity levels in fishotherwise identical fish, but provides a means number five, with low ""Zn but higher 32P of formulating alternate hypotheses regarding specific activity, suggest that this fish hadtheir recent history.These hypotheses can recently accumulated itsradioactivity.The then be tested by more conventional fisheries other 14 fish, which because of the higherinvestigations, thus leading to a better un- 85Znconcentrations seem to have been influ- derstandingoftheColumbia Riverfish enced by Columbia River radioactivity forpopulation. a longer period, appear toconstitute two groups. There is a group of nine individuals ACKNOWLEDGMENTS (Nos. 9, 10, 11, 12, 16, 19, 20, 21, and 24) This study is part of a thesis submitted by with low 32P specific activities(below 1.1 the senior author toOregonState University nCi/g P), but higher C5Zn concentrations (26 in partial fulfillment of a Master of Science to 97 pCi/g dry wt).This would indicate Degree. The work was supported by a that these fish accumulated their radioactivity Federal Water Pollution Control Administra- at anearlier date and that subsequently the tion traineeship and the U. S. Atomic Energy 32P levels had decreased for some reason CommissioncontractnumberAT(45-1) (e.g.,reduced winter food consumption). 1750,Pub. No.RLO-1750-84.We are grate- The other group of five fish (Nos. 8, 14, 18,ful to Drs. William 0.Forster,Norman H. 22, and 23) have both higher 32P specific Cutshall,and Mr. Lauren Larsen for advice activities (2.5 to 5.1 nCi/g P) and high 65Zn and analyticalassistance.We also thank the concentrations (55 to 94 pCi/g dry wt). This biologistsoftheU.S.National Marine second group of fish with higher 32P specific Fisheries Service for collecting samples. activities and encompassed by the arbitrary boundaries, must have accumulated their 32P LITERATURE CITED at a laterdate than the group with lower 32P DAVIS, J. J., AND R. F. FOSTER.1958. Bioaccumu- specificactivitieswhichareoutsidethe lationofradioisotopesthrough aquaticfood boundaries.Speculations such as abnormal chains. Ecology 39(3): 530-535. FOSTER, R. F., AND D. MCCONNON.1962.Relation- feeding, seasonal equilibrium changes, migra- ship between the concentration of radionuclides tion from further upriver or into tributary in Columbia River water and fish, p. 216-224. InClarence M. Tarzwell (ed.), Biological prob- streams, oroverwintering can be made to lems in water pollution. Cincinnati, Ohio, U. S. explain the differences, but these would have Robert A. Taft Sanitary Engineering Center. to be verified. U. S. GEOLOGICAL SURVEY. WATER RESOURCES DIVI- SION. 1968.Average monthly discharge for the The best application of the graphical model 15-year base period, water years 1953-67 inclu- may be to identify, within a' population of sive, at selectedgauging stationsand change of seemingly identical fish, groups of fish which storagein certainlakes and reservoirs in the Pacific Northwest. Data for stations in British apparently have had quite different recent Columbia furnished by Water Survey of Canada. histories. Boundaries for the graphical model Washington, D.C.1 sheet (Table no. 1, suppi.). are variable, being highly dependent on sea- WALLIS, J.1968.Recommended time, size, and age for release of hatchery reared salmon and steel- son, reactor operation, species of fish, and head trout.Clackamas, Oregon: Fish Commis- locationalong the river. Opportunity for an sionof Oregon, Research Division.61 p. WATSON, D. G., C. E. CUSHING, C. C. COUTANT, AND environmental uptake study is also not always W. L. TEMPLETON.1969.Effect of Hanford available. The model will, therefore, be of reactor shutdown on Columbia River biota, p. 291-299. In Daniel J. Nelson and Francis C. greatestvalue when analysis of a single large Evans (eds), Symposium on radioecology. U.S. population shows general trends making it Atomic Energy Commission.Conference-67503. 60

reprinted from: RADIONUCLIDESIN ECOSYSTEMS Proceedings of the Third NationalSymposium on I?adioecology,D. J.Nelson,Editor CONF-710501 (1973)

RLO-1750-86

SPECIFICACTIVITY OFIRON-55IN PACIFIC SALMON

C. DavidJennings'and Charles L. Osterberg' Department of Oceanography Oregon State University, Convallis

ABSTRACT

Fallout 55Fe from the nuclear tests of 1961-62 was measured in Pacific salmon during 1964-67. and from these measurements the half-time for the loss of SsFe from the mixed layer of the North Pacific Ocean was estimated to be 11.5 months. On the assumption that salmon accumulate 55Fe and stable iron in a ratio corresponding to the specific activity of 55Fe in the seawater in which they feed, estimates are made of total 55Fe in the North Pacific Ocean. Circumstantial evidence suggests that sSFe from fallout isin a form which is more readily available to marine organisms than the stable particulate iron in the ocean; thus fallout 55Fe may not be an adequate tracer for stable particulate iron in the ocean.

INTRODUCTION 1961-62 were at Johnston Island (17° N, 169° W) and Christmas Island (2° N, 157° W) (Peirson and Cantbray Iron-55 from fallout, formed predominantly by (n,7) 1965). At the latitude of the Russiantests, debris reactions on stable $4 Fe in bomb hardware, constituted wouldbe carriedin an easterlydirection bythe a significant portion of the total radioactivity in the prevailing westerlies, whereas debris from the American mixed layerof the North Pacific Ocean after the tests in thePacific would be transported to the west by nuclear weapons testing of 1961--62. Little is known, the trade winds and the North Equatorial Current. however, about the oceanic distribution of 5 5 Fe despite Thus, fallout from the Russian tests was carried into its relative importance as a fallout radionuclide. our study area,theEastern North Pacific Ocean, Iron-55 inthesea can he more easily studied by whereas fallout from the American tests was carried measuringitinmarineorganismsthan bydirect away from our study area. measurement in seawater. because concentrations are In our study, the specific activity of 5 5 Fe (defined as usually much higherinorganisms thaninwater. theratio activity SSFe;weight totaliron) in ocean Furthermore, collections of marine organisms available organisms was 250 times higher in 1964 than just prior from pastyears preserve the record of "Fe in the sea to resumption of testing in 1961 (Rama, Koide, and during those years. The 2.7-year half-life of 5 5 Fe allows Goldberg1961). So there did not appear to be a a reasonable time for analysis of these older samples reservoir of 55Fe in the ocean remaining from earlier before the "Fe decays below the limits of delecta- nuclear tests even though 55Fe has a 2.7-year half-life. bility. For these reasons our assessment of "Fe in the A large increasein "Fe from tropospheric fallout mixed layer of the North Pacific Ocean has been from appeared in the North Pacific in a short time after the analyses of Pacific salmon, which contained the highest tests, andstratospheric fallout contributed an ever concentrations of55Fereportedfor any organism decreasingamount. We conclude, therefore, that fallout examined(Palmer, Beasley, and Folsom 1966). from the Russian nuclear tests of 1961 -62 gave rise to relatively high concentrations of "Fe in the North SOURCE OF IRON-55 Pacific Ocean and thatthe contribution inrainfall IN THE NORTHPACIFIC OCEAN occurring after thisinitiallarge input in 1962 was The "Fe that we measured in this study appears to significantly less. have been introducedinto the NorthPacificOcean by the Russian nuclear tests of 1961-62. The Russian test TREATMENT OF SAMPLES sites were in the Arctic (75' N, 55° E) and in Central In this study we measured 55Fe in viscera from three Asia (52° N, 78° E), and the American test sites of species of salmon provided by commercial fishermen who fish the Eastern North Pacific Ocean. Our samples 1. Present address Department of Natural Sciences and were caught ateight stations between Bristol Bay, Mathematics, Oregon College of Education, Monmouth. Alaska, and Eureka, California. Samples were preserved 2. Present address: Division of Biology and Medicine, U.S. in formalin and shipped to the laboratory, where they Atomic Energy Commission, Washington, D.C. were dried at 100 C and ashed at 600 C for 24 hr in a

703 61

704

muffle furnace. The ached samples were dissolved in 6 activity of seawater by an amount determined by the At hydrochloric acid; analiquot ' was removed for decay of 55Fe as it moves tip the food chain. Because spectrophotometric determination of stable iron, and ssFe hasa 2.7-year half-life, itis reasonable to expect the remainder of the sample was used to measure 5 5 Fe. that little radioactive decay occurs and that the specific Iron-55 was measured by first extracting iron from activity of SSFe in oceanic organisms gives a good the dissolved sample with 10; Alanline-336 (available estimate of the specific activity of the seawater in from GeneralMills.ChemicalDivision,Kankakee, which they 'ive. Assuming marine organisms concen- Illinois: trade names referred to in this paper do not trate ssFe at the same specific activity as occurs in imply endorsement of commcicial products) in xylene, seawater, the product of the specific activity of oceanic electroplating the iron by the method of Maletskos and organisms and the average concentration of iron in the Irvine (1956), and counting the K x rays of the 55Mn ocean (10 pg/liter, Goldberg 1965) approximates the daughter of "Fe in a Las flow proportional counter, average concentration of 55 Fe in seawater at the time anticoincidence shielded by two 24 X 10 cm Nal(TI) the organisms accumulated the isotope. crystals. The method has been described indetail Iron-55 in the ocean was first estimated in this way in ssFe in elsewhere (Palmer and Beasley 1967). 1961 on the basis of the specific activity of tuna(Rama, Koide, and Goldberg1961), and we SPECIFIC ACTIVITY OF IRON-55 IN SALMON estimated55 Fein the North Pacific Ocean similarly for 1964, 1965, and 1967 on the basis of our analyses of The specific activities of "Fe in salmon in 1964. salmon (Table 2). 1965, and 1967 (Table 1) represent averages of more than ISOseparate analyses of samples from eight Table 2. Estimate of sSFe in the North Pacific Ocean stationsintheNortheast Pacific Ocean. The most from specific activity of 55Fc in salmon complete set of samples was from Bristol Bay. Alaska, ssFe at collection time 5SFe corrected to 1962 Year so averages of measurements in years after 1964 were (mCi/km2) (mCi/km2) normalized to the specific activity of ssFe in Bristol Bay salmon. 1961 0.037 x 103 A 250-fold increaseinspecificactivityof "Fe 1964 9 x 103 I S X l03 occurred between 1961 and 1964 which reflects the 1965 3 X 103 7 X 103 large input intothe oceanof ssFe from the 1961-62 1967 0.5 X 103 2 X 103 nuclear tests. The decrease in specific activity of 55 Fe in each succeedingyear after 1964 willbe discussed 'Calculated from tuna data of Rama, Koide, and Goldberg later. (1961).

Table 1. Specific activity of "Fe in Pacific salmon Calculations of the's Fe inventory of the North Average specific activity PacificOcean were made on the assumption that Year of collection (pCi/g Fe) salmon obtain their iron from the mixed layer of the ocean wherethey feed.We assumed a mixed layer 100 1%1 0.037' m thick and an average iron concentration in the ocean 1%4 9.1 of 10 pg(liter and corrected all ss Fe measurements for 3.4 1965 decay back to 1962,when the original large 1967 0.5 radioactive input of 5 5 Fe in the ocean took place. From tuna data ofRama, Koide, and Goldberg (1961). The assumption that radioisotopes deposited as fallout on the ocean behave in a manner similar to their stable counterparts is notnew.It forms [lie basis for ESTIMATE OF IRON-55 IN THE OCEAN FROM THE marine geochemical tracerstudies.biological uptake SPECIFIC ACTIVITY OF IRON-55 IN SALMON studies, and the specific activity concept for prediction Oceanic organisms concentrate ssFe from their envi- of maximum permissible concentrations of radioactivity ronment and thus facilitate measurement of5 5 Fe in the inseawater(NAS-NRC1962).If the chemical or ocean. If we assume that organisms accumulate 55 Fe physical forms of the stable and radioactive isotopes of and other isotopes of iron in the same ratio as they an element differ, however, it is doubtful that knowl- occur inseawater, we would expect the specific activity edge of the role of the radioisotope can be extended to of "Fe in the organisms to be less than the specific the stable isotope. 62

705

Some studieshave suggestedthat certainradio- compare specific activities of salmonand water col- isotopes can behave differently from their stable iso- lected simultaneously, it would be more reasonable to topiccounterparts in seawater. Schelske et al. (1966) compare the specific activityof ssFe in salmon with found the specific activity ofso Mn to be higher in thespecificactivity of "Fe in seawater collected estuarinescallopsthaninthesurrounding water. during the time of the salmon's period of most rapid Slowey et al. (1965) suggested the possibility that the growth (from one to six years before maturity, depend- chemical form in which 5Ninreachestheseais ing on the species). Unfortunately, such measurements different from that for stable manganese and, thus, that are not available for seawater, butconsidering that the 54Mn isnot in equilibrium with stable manganese. input of fallout5 s Fe to the oceans has been decreasing Robertson etal. (1968) showed thatthespecific from 1963 to the present, it should not he surprising to activitiesof 60Co and 6SZn were generally several find mature salmon with a specific activity higher than ordersof magnitude higher in marine organisms than in the watersfrom which they were caught. seawater, but that the specific activities of 137Cs in Although higher specific activities of a radioisotope in marine organismsand seawater were essentially the marineorganisms than in their seawater environment same. Also, the specific activity of "Fe in salmon must not be taken as conclusive evidence forselective caught between 1964 and 1967 was 1000 to 10.000 uptakeof the radioisotope, it does provide circumstan- timeshigher than the specific activity ofssFe in the tial evidence. A growing body of this type of evidence particulate fraction of seawater off the northern Cali- suggests thats s Fe from fallout is in a form which is fornia coast (Palmer et al. 1968) and off the Oregon more readily available to marine organismsthan the coast (Robertson et al. 1968) in 1968. stable particulate iron in the ocean. Some comparisons A lower specific activity for a radioisotope in water in Table 3 support this hypothesis. Calculation of the than in amarine organism suggests preferential uptake s s Fe reservoir in the ocean on the assumption that the of the radioisotope over its stable counterpart. Caution specific activity of ss Fe in salmon equals the specific s S Fe should beexercised, however,in drawing such a activity of in seawater produces estimates of5 s Fe conclusion. Because of local variations in fallout and in theNorth Pacific Ocean which are over 100 times stable element concentrations, the specific activity of a higher than the contribution of ssFe in rainfall at sample of seawater collected in one location should not Westwood. New Jersey. Part of the discrepancy prob- be expected to equal the specific activity of a marine ably representsreal differences in fallout between the organism from another location. Furthermore, samples two locations. The North Pacific Ocean is much closer of seawater and marine organisms collected simultane- to the Russiantest sites than is Westwood. Alaskan ously may have specific activities at variance with one lichen,also closerto the test sites than is Westwood, fit another, because the organism obtained its radioactivity into this scheme with intermediate values of fallout at anothertime or in a different body of water. It has ssFe. Likewise,"Fe inthe North Atlantic Ocean in been suggestedthat salmon retain most of the iron they 1965, calculated from our analyses of false albacore in take up(Palmer and Beasley 1967). Thus, rather than the same way asthe North Pacific Ocean data, was

Table3.Comparisonsof SSFein theNorth Pacific Ocean calculated from ssFe insalmon.SSFe in the North Atlantic Ocean calculated from SSFe in falsealbacore,SSFe inlichen,and SSFein rainfall

Iron-55 inthe North Iron-55 in the North Iron-55 in rainfall Iron-55 in Alaskan Iron-55 in rainfall Pacific Ocean from Atlantic Ocean from atWestwood. N.J.,* Year specificactivity of specific activity of lichen, decay in France," decay decay corrected to corrected to 1962 salmon viscera, false albacore viscera, corrected to 1962 1962 (mCi/km2) decay corrected to 1962 decay corrected to 1962 (mCi/km2) (mCi/km2) (mCi/km2) (mCi/km2 )

1963 249 54 1964 15,000 108 30 1965 7,000 700 48 t6 1966 231 18 1967 2,000 187

*From Hardy and Rivera (1968). ''From lloang ct al. (1968). 63

706 aboutone-tenth of the "Fe in the North Pacific Ocean this assumption is not strictly correct because of the in 1965. Finally, stillless 55Fe Was measured in rainfall fallout contribution each year, but in the North Pacific in France, with a greater distance of travel from the Ocean the relatively large increment of fallout immedi- Russiantest sites, than in rainfall in Westwood. ately after testing makes this a workable assumption. The discrepancy between "Fe in the ocean calcu- A semilog plot of s"Fe in the ocean versus time lated from measurement of "Fe in fish and "Fe resultsin a straight line which suggests a first-order measured in rainfall at Westwood and in France appears reaction as the mechanism for removal of "Fe (Fig. I). to be too great to be explained solely on (lie basis of The equation for the plot of Fig. 2 is derived from proximity to the test site, and thus provides circumstantial evidence that fallout "Fe and stable iron in the -dA/dt=k,A, (1) ocean do not behave in the same way. The discrepancy can benoted particularlyinthe15-fold difference giving between SSFe inthe North Atlantic Ocean and in rainfall at Westwood in 1965 and theI I-fold difference -k, between 5 SFeinthe North Pacific Ocean and in log A = 2.303t+logA0, (2) Alaskan lichen in 1967. where LOSS OF IRON-55 FROM THE A = radioactivity of 55 Fe per unit area in the ocean OCEANIC MIXED LAYER at t, When the estimates of SSFe in the North Pacific A0 = radioactivity of 55 Fe per unit area in the ocean Ocean for 1964, 1965, and 1967 are all decay-corrected att=0, to 1962, an annual decrease occurs which indicates k,=rateconstant forremoval of 55 Fe from the reduced availability of "Fe to salmon due to some mixed layer of ocean. process other than radioactive decay. To calculate how long " Fe remains available to organisms in the surface From the slope of the line described above by Eq. (2), layer of the ocean, we assumed that all the 55 Fe in the k,was calculated to be 0.0603 month-' and the North Pacific Ocean from 1964 to 1967 was deposited half-time for the removal of "Fe from the surface in a shorttime following the nuclear tests. Of course, layer of the ocean to be 11.5 months.

0.1 1964 1965 1966 1967 0 1 2 3 4 YEAR OF SAMPLE COLLECTION TIME AFTER FALLOUT DEPOSITION ( YEARS )

Fig.1. Decrease of SSFe in the mired layer of the North Fig. 2. Decrease of SSFe in the ocean calculated on the Pacific Ocean from 1964 to 1967 based on measurements of the assumption that Swke's law accounts fox all the loss of SSFe specific activity of 55 Fe in salmon. from the mixed layer. 64

707

POSSIBLE LOSS MECHANISMS mean half-time of 13.6 months is near the half-time of 11.5 months observed in the salmon data. The settling of fallout particles by gravitational forces A second possible explanation for the decrease of as described by Stoke's law is one way in which s s Fe ssFe measured in salmon may he found in the decrease might be lost from the mixed layer of the ocean. To of s s Fe in rainfall. If fallout s' Fe becomes unavailable calculate the loss of fallout from the mixed layer as to salmon in a short time after its introduction to the predicted by Stoke's law, one must know or assume the ocean,the amount of ssFe insalmon would be fraction of thetotalradioactivitypresentin each regulated by the amount of ssFe in rainfall, and the particlesize fraction.It would be most suitable to decrease of ssFe in salmon should correspond to the assign a size distribution to fallout particles on the basis decrease of s s Fe in rainfall. Any residual s s Fe in the of actualsampling of the nuclear tests in question, but ocean should have the effect of lowering the apparent since these data were not available, we assumed the rate of decrease ofs s Fe in salmon relative to rainfall. particle size distribution found by Sisefsky (1961) to be Comparisons by Palmer et al. ( 1968), however, show a applicable. We chose the six size ranges shown in Table higher rateof decrease of "Fe in salmon than in 4 and calculatedsettlingvelocities for each mean rainfall.Thisdiscrepancy canbe explainedif,as particle size. suggestedearlier, the fallout"Fein the North Pacific Calculations of settling for each particle size were Ocean front1964-67 resultedlargely from tropo- made on thebasis of complete mixing of the upper 100 sphericfallout, with stratospheric falloutmaking a in every 24hr (Wooster and Ketchum 1957) from the relatively minor contribution. Stoke's law equationforthelimiting velocity of Other loss mechanisms which may be operating to spherical particles in a fluid, reduce the concentration of's Fe in the mixed layer of the NorthPacific Ocean include horizontal or vertical 2 diffusion and dilution by horizontal or vertical mixing. vlim =2 g(p- po)r2 (3) Itislikelythatthese and other processes allare operating simultaneouslytoproducethe observed where result. 11= viscosity of seawater = 1.39 X 10-2, CONCLUSIONS g = 980 em/sees, Pacific salmon are concentrators of 55 Fe in the ocean p = density of particles = 3.70 (Benson et al. 1967), and, thus, contain high specific activities of ssFe. Whether they achieve the same specific activity as the po = density of seawater = 1.03. seawater in which they feed, so that they preserve the A semilog plot of ssFe remaining in the mixed layer record of the ssFe reservoir in the sea, has not been after Stoke's law settling has been accounted for versus firmly established. It is possible that fallout's Fe is in a time is shownin Fig. 2. The half-time for the removal form different from the form of stable particulate iron of "Fe varies from 8.2 to 19.0 months, so that the in seawaterand that salmon or organisms on which salmon feed selectively filter fallout ssFe from seawater. However, evidence for isotopic discrimination or Table 4. Percent of total radioactivity in each of six mean selective uptake by salmon of "Fe Fe from seawater is sizes of fallout particles and their associated velocity circumstantialonly, and much of the observed dif- of settling in the ocean predicted by Stoke's law ference could be caused by geographical variations in Mean Initial fallout deposition and by real differences between the Size Settling particle total range" velocity specific activities of the seawater when the salmon size W) (m/day) radioactivity* obtained their ssFe and the specific activity of sea- W) in size range (%) water when it was sampled. In the event of renewed

0.1--0.5 0.30 0.0081 1.0 nuclear testingin the atmosphere, simultaneous pro- 0.5-1.0 0.75 0.051 8.7 gramsof sampling seawater and marine organisms 1.0-1.5 1.25 0.141 30.6 should be undertaken to resolve the present dilemma. 1.5-2.0 1.75 0.276 29.2 2.0-3.0 2.30 0.477 14.8 ACKNOWLEDGMENTS 3.0-4.0 3.50 1.11 15.7 We wish to thank II. E. Palmer and R. M. Bernard for *Adapted front Sisefsky (1961). their assistancein counting thesamples,and Norman 65

708

Kujala and W. C. Hanson for help in obtaining samples. Palmer, H. E., J. C. Langford, C. E. Jenkins, T. M. This researchwas supported by USAEC Contract Beasley, and J. M. Aase. 1968. Levels of iron-55 in AT(45-I )1750 and PHS Training Grant STI-WP-1 11-01. humans, animals, andfood, 1964-1967. Radiol. (RLD-1750-86) Health Data Rep. 9: 387-90. Peirson. D.H., and R. S. Cambray. 1965. Fission product fallout from the nuclear explosions of 1961 LITERATURE CITED and 1962. Nature 205: 433-40. Rama, Minoru Koide, and E. D. Goldberg.1961. Benson, P.A., M. W. Nathans, A. Amos, and L. Iron-55 from nuclear detonations. Nature 191: 162. Leventhal. 1967. The density of fallout particles from Robertson, D. E., W. O. Forster. H. G. Rieck,and J. C. airbursts. Health Phys. 13: 1331-35. Langford. 1968. A study of the trace element and Goldberg. E. D. 1965. Minor elements in seawater, pp. radionuclide behavior in a northeast Pacific Ocean 163-96 in: Chemical Oceanography, vol.1, J. P. ecosystem 350 milesoff Newport, Oregon, pp- Riley and G. Skirrow (Eds.). Academic Press, New 92-108 in:Pacific Northwest Laboratory Annual York. Report for 1967 to the USAEC Division of Biology Hardy,E. P., Jr., and J.Rivera. 1968. Fission product and Medicine, Vol. 11: Physical Sciences,Pt.2: and activationproduct radionuclidesin monthly Radiological Sciences. BNWL-715. deposition at selected sites, pp. B-1-B-22 in: Appen- Schelske, C. L., W. D.C.Smith,and Jo-Ann Lewis. dix to Healthand Safety Laboratory. Fallout Program 1966. Radio-assay, pp. 12-14 in: Annual Report of Quarterly Summary Report. January I, 1968. HASL- the Bureau of Commercial Fisheries Radiobiological 184. New York. Laboratory, Beaufort,N.C.,fortheFiscal Year Hoang,C. T., J.Servant, and J.Labeyrie.1968. Ending June 30, 1965. Circ. U.S. Fish. 1Vildl. Serv., Evaluationde la retombee mondiale du "Fe a la 244. suitedesessaisnucleaires dins ('atmosphere des Sisefsky, Jan.1961. Debris from tests of nuclear annees 1961 et 1962. Health Phys. 15: 323-32. weapons. Science 133: 735-40. Maletskos, C. J., and J. W. Irvine, Jr. 1956. Quantitative Slowey, J. F., DavidHayes. Bryan Dixon,and D. W. electrodepositionof radiocobalt,zinc,andiron. Hood. 1965. Distribution of gamma-emitting radio- Nucleonics 14(4): 84-93. nuclides inthe Gulf of Mexico. pp. 109-29 in: National Academy of Sciences - National Research Marine Geochemistry. D. R. Schink and J. T. Corless Council. 1962. Disposal of low-level radioactive waste (Eds.). Occ. Publ., Narragansett Marine Laboratory, into Pacific coastal waters. NAS-NRC Publ. No. 985, Graduate School of Oceanography, Univ. Rhode Washington, D.C. 87 pp. Island, No. 3. Palmer, H. E., and T. M. Beasley. 1967. Iron-55 in man Wooster, W. S., and B. H. Ketchum. 1957. Transport and the biosphere. Health Phys. 13: 889-95. and dispersalof radioactive elements in the sea, pp. Palmer, H. E., T. M. Beasley, and T. R. Folsorn. 1966. 43-51in: The Effects of Atomic Radiation on Iron-55 in marine environment and in people who eat Oceanography and Fisheries. NAS-NRC Publ. No. ocean fish. Nature 211: 1253-54. 551, Washington, D.C. 66

reprinted from:' RADIONUCLIDES IN ECOSYSTEMS Proceedings of the Third National Symposium on ladioecology,D. J. Nelson, Editor CONF-710501 (1973)

RLO-1750-88

EFFECT OF GAMMA IRRADIATION ON THE REPRODUCTIVE PERFORMANCE OFARTEM/AAS DETERMINED BY INDIVIDUAL PAIR MATINGS' Robert L. Holton' and Charles L. Osterberg Divisionof Biology and Medicine U.S. Atomic EnergyCommission,Washington, D.C. William 0. Forster Puerto Rico Nuclear Center College Station, Mayaguez, Puerto Rico

ABSTRACT The brine shrimp, Artemia,was used as an experimental organism to study the effects of 60Co gamma irradiationon the reproductive performance of an animal species. The total reproductive ability of the brine shrimp was fractionated into various components, and the effect of irradiation on each of thesecomponents was then determined by studies of reproductive behavior in individual pair matings. In this study the components identified were the number of broods produced perpair,the number of naupliivoided perbrood,the survival of nauplii to sexualmaturity,the number of mature adults producedper brood, and finally the number of mature adults produced per pair. All component parameters of total reproductive performance were shown to be affected by irradiationwithin the range of doses used in thisexperiment. However,the number of broods per pair was shownto be the factor most sensitiveto irradiation,with effects noted at doses of 900 rads and above.

INTRODUCTION from this simple. We must not only know the effects of these various doses of irradiation upon the reproductive Thereisa sizable amount of literature which is ability and physiological functioning of the population, concerned with the effects of ionizing radiation on but we must also know the degree to which a particular physiological processes as well as with genetic change plant or animal species concentrates the various radio- and the subsequent appearance of mutations in future nuclides, where they are located in the organism. and generations. The recent publication by Grosch (1965) what dose these nuclides will deliver to thegerminal summarizes much of this research. The emphasis in the tissue as well as to important somatic tissues which present study is different. In this case the irradiation is mightaffectthephysiologicalfunctioning of the consideredas anecological factor, and its effect on the organism in a competitive situation. reproductive performance of a laboratory population is An experiment has been performed to evaluate the evaluated. Only by successfulreproduction can a effects of acute doses of gamma rays, from a "Co species maintain its place in the ecosystem. source, upon the reproductive ability and hence on This study was initiated to add to our understanding population maintenance of laboratory populations of of the effects of various doses of gamma irradiation the brine shrimp,Artentia.The brine shrimp was chosen upon the reproductive ability of an animal population. as the experimental organism for several reasons. First, Such knowledge is essential if we are to understand the it was considered desirable to extend population radio- effects of present and future doses of irradiation upon sensitivity studies of crustaceans. Second, the brine the various populations and hence on the, production shrimp proved to be a very tractable laboratory animal and evolution of various ecosystems of our modern that could easily be reared through many successive world. However, in the environment the situation is far generations. The brine shrimp is an especially favorable animal for these studies, sinceitis possible to frac- I. Informationcontained in this article was developed during the course of work under fellowship number 1-1-1-WP-30, tionate the reproductive performance into several com- 166-01 of the USPHS and 1-1 I-WP-30. 166-02 of the Federal ponents and to study the effect of irradiation on each Water Pollution Control Administration. parameter. Finally, the attention was focused on this 2. Presentaddress: Department of Oceanography, Oregon animal by the work of Grosch (1962, 1966) in which he Slate University, Corvallis, Oregon 97331. studied the effects of the accumulation of the radio-

1191 67

1192 nuclides 6 5 Zn and 32p upon the reproductive behavior brief review of the life cycle ofArtemia.A more ofArtemia.This research allows some comparison detailed presentation of the life cycle will be found in between the effects of incorporated radionuclides and Holton (1968). the effects of irradiation from an external source. Upon hatching,this genus emerges as atypical In particular, this series of experiments was designed nauplius bearing the customary three pairs of append- toestablishdosagelevels,inrads,at which the ages. In successive instals we observe the development reproductive ability of Artemia was affected and to of the thoracic appendages, modification of the body determine the dose required to reduce the reproductive shape, and specialization of the head appendages. In ability of the population to zero and thus fix the level particular at about the eighth instar we can detect a at which a population would be unable to reproduce sexual dimorphism manifest in the structure of the itself and hence go to extinction under laboratory second antennae. In the female the second antennae conditions. which were used for locomotion in earlier stages are reduced greatly in size. In the male the structure of the METHODS AND MATERIALS second antennae is increased in size and modified in structureto be used to grasp the female during The organism. The genus Artemia is a member of the copulation. By the ninth instar the modifications are order Anostraca, subclass Branchiopoda, class Crustacea distinct and the sexes easily distinguishable. of the phylum Arthropoda (Barnes 1963). Each of the After themale attains sexualmaturity,he grasps the members of the Branchiopoda has an epipodite on the female duringcopulation,with his enormous specialized thoracic appendages which serves as a gill, hence the secondantennae. The female violentlyresiststhis name Branchiopoda, meaning "gill feet." The Anostraca action, but a successful male secures a position dorsal to are called "fairy shrimp." They have no carapace and and partly behind the female, with his antennae locked the compound eyes are stalked. Other members of the around her body just in front of the ovisac. The two order typically inhabit temporary freshwater ponds, animals are firmly united and swim about in this while the genusArtemiais found in both temporary fashion. At intervals the male curls forward the hind and permanent ponds of high salinity. part of his body,attempting to insert one of the two Although laboratory experiments show thatArtemia male organs into the uterus. The female does not permit survives and reproduces for many generations in sea- this unless she has recently completed a molt, successful water, itis never found in this environment in nature. copulation being restricted to this period. However, itis found widely distributedin more saline The reproductivesystemof thefemale consists of two waters in many parts of the world. It is apparently able ovaries, two oviducts, and a ventral uterus. The follow- to survive only where members of higher trophic levels ing sequence of events occurs in the adult female are eliminated by the increased ionic concentration of (Bowen 1962). The female expels from theuterus an these waters(Carpelan 1957). egg generation, the process taking 2 to 10 hr. Then she Due to the taxonomic problems cited by Holton molls in a few seconds, and the next egg generation (1968) we will refer to the organism used only by the passes from the ovaries to the oviducts in less than 2 hr. generic nameAnemiainthis paper. However, the Copulation appears to be effective only while the eggs particular organisms used were from the population remain in the oviducts, with fertilization occurring as which inhabits the Great Salt Lake, Utah, in the United they pass from the oviducts into the uterus. Shortly States. The particular cysts used were collected in after the eggs enter the uterus, the shell glands may August 1965 by the Brine Shrimp Sales Company of begin to secrete the dark shell material which surrounds Hayward, California. All experimental work reported the cysts. was conducted between September 1966 and June Lochhead (1941) andBowen(1962) have shown that 1967; hence the cysts were between one and two years under ideal conditions the shells are not formed and the of age during the course of this work. cysts hatch to yield free living nauplii beforethey are This population is a diploid amphigonic population expelled from the uterus. Hence we have a viviparous (Bowen 1964) which is well known due to its extreme type ofbirth. However, in some cases a thin shell is abundance in Great Salt Lake. Cysts are harvested formed and the cysts which are expelled typically hatch regularly and sold commercially to be hatched as food in about one day. These cysts are often called "summer for aquarium fishes. eggs." A third possibility is the formation of the thick The life cycle.Itwill be possible to discuss the shelled "winter eggs," which are capable of surviving for reproductive performance of this organism only after aa long time under adverse conditions. 68

1193

The life cycle is advantageous fur studying reproduc- Individual pair matings were made in approximately tive performance, since itis possible to follow several 500 ml of culture water (1-pt polyethylene freezer parametersin a single experiment. Each pair produces cartons). Each pair mating was examined daily, and several broods during their lifetime, and therefore it is when young were noted, the parents were transferred possibletoquantitatively determine the effect of into a new culture container and the young were raised irradiation on the number of broods produced per pair, to maturity in the original container. Nauplii were the number of nauplii voided per brood, the number of counted on the third day after they emerged, returned nauplii voided per pair, the survival of nauplii to sexual to the culture container, and counted again and sexed maturity, the number of mature adults produced per at maturity. All animals were fed daily a standardized brood, andfinallythe number of mature adults suspension of brewer's yeast in distilled water. produced per pair. Conditions of irradiation. The ionizing radiation used Culturemethods. All cultures were maintained under in all experiments consisted of gamma rays (1.17 and constant illumination in the laboratory. The laboratory 1.33 MeV) from "Co. The animals were irradiated-in used had no outside windows, and it was found that the 10 ml of fresh culture water held in 50-m1 polyethylene animals reactedto the sudden turning on of the room test tubes. Five of these test tubes could be irradiated at lights after being in total darkness by several minutes of one time in the specially constructed polyethylene very rapid random movement before settling down to a holder. In the cases when not all of the five tubes were normal swimmingpattern typically exhibited in the being used to irradiate animals, the unused tubes were light. Duringthe course of the experiments the temper- placed in the tube holder with 10 ml of culture water in ature ofthe laboratory was maintained within the range order to maintain a constant geometry of irradiation. of21.0+0.7C. During the course of the irradiation, the irradiation Since all experimental work in this study was con- chamber was continuously perfused with air. ductedusingArtemiacysts from Great Salt Lake and Inallexperiments, parallel control cultures were sincein their normal environment this population is maintained. These animals were subjected to the same subjected to salt concentrations of several times that of handling as the irradiated animals and were placed in seawater,the question of the proper salt concentration the irradiator for a time equal to the longest irradiation to use in a laboratory study arose. A small experiment time used in that experiment. However, for the control was designedto help resolve this question. Several cultures the safety shieldis not removed, and hence groups of animals were reared in seawater and in they receivevirtuallyno irradiation. A secondary seawater with variousamounts of sodium chloride control which was subjected to as little handling as added, of from 50 to 150 g/liter. When mature, the possible was also maintained to help assess the extent of shrimp wereirradiated at several different doses and damage from handling. This secondary control showed their survivalwas evaluated. The data are graphed in Fig. 1. It should be noted that at all levels of irradiation the 100r survival wasbest in the seawater with 50 g of added sodium chloride per liter, and this waterwas used as a standard culturewater. It would seem likely that this 80 culturewater would most nearly approximate the environmentalconditions of Great Salt Lake, in which 60 this population evolved, of any of the various waters Orods tested. Itis interesting that both Bowen (1962) and Grosch (1966) have also adopted culture water with 50 40 g of NaCl added to each liter of seawater for their work with Artemia, although they used a different criteria for Z5 25,000 rods selection of this culture water. 20 All cysts were hatched in seawater, with hatching 4,000 rods being complete in 48 hr after hydration of the cysts. At 01 I 1 I 1 48 hr after hydration, the newly hatched nauplii were 0 +50 +100 +150 transferred with a large pipette to the standard culture S£AWAT£R (g. NaCi/titer) water and remained in this water during the course of Fig. 1. Percent of young adult Artemia surviving for 25 days the experiments. at various doses of irradiation in four different culture waters. 69

1194 that the handling during irradiation did not affect the beforetheresults would be obscured by broods survival of theArtemia. produced from oocytes and spermatocytes which were formed before the irradiation. A second preliminary experiment resolved the stage of maturity for irradia- RESULTS AND DISCUSSION tion necessary to satisfy the above requirements. It was found that irradiation at either the 10th or I Ith instar This experiment was conducted to study the effects described by Heath (1924) would irradiate germinal of an acute dose of gamma irradiation, from the "Co tissue at a time of maximum radiosensitivityand would source, upon the reproductive performance of Artemia still be before any radioresistant germinal products were in individual pair matings. During the course of a series formed. This also worked out to be the 22nd or 23rd of preliminary experiments, it became very clear that day after the hydration of the eggs under thestandard the stage of the life cycle at which theArtemiawere condition employed in this laboratory. irradiated had an important effect on the subsequently The animals forthis experiment were hydrated, observed reproductive pattern. hatched, and transferred according to thestandard Irradiation of the early nauplii stages in the lifecycle technique as previously described when discussing the produced marked effects on reproductive ability, of conditions of culturingArtemia.After hatching they those which survived the irradiation, but only at were placed in an 8-gal aquarium tank for development. relatively high doses of approximately 2000 rads and Twenty-two days after hydration,the animals were above. However, the irradiation of these nauplii at such removed from the aquarium with the aid of a fish net, doses produced detectablesomaticdamage. The irradi- and a sample was mounted and examinedunder the ated nauplii showed a high mortality rate, and such microscope to determine the stage of development. It doses also slowed their rate of development to adult- was found that the development was very uniform with hood. This lowerrateof development effectively a few animals in the I 1 th instar and between 80 and eliminated irradiation of nauplii from this study, since 90% in the 10th instar. The animalswere separated into animals receiving different doses became reproductively groups of 80 unsexed animals. One group of 80 such mature at varying times, and hence the reproductive animals was irradiated in 10 ml of culture fluid under performance was confounded with developmental rate the standard conditions of irradiation as previously in such a way as to make analysis of any data difficult. described, at each of the following doses: 0, 300, 600, As theArtemiaapproached maturity, their reproduc- 900, 1200, 1500, 1800. 2100, 2400, 3000, 4500, and tive ability became increasingly more sensitive to the 6000 rads. The smaller preliminary experiments pre- effects of gamma irradiation. At the sametime as they viously mentioned had indicated that thedoses at were becoming reproductivelymore sensitiveto irradia- 300-rad intervals to 2400 rads should bracket the entire tion, they were showing greatly increasedresistance to range of irradiation within which reproduction would somatic damage, as evidenced by loweredmortality at be possible. The three higher doses wereincluded in the range of doses used in this experiment. Accepting case the animals irradiated at 2400 radsshould show the general principle that cells aremost sensitive to any reproductive ability. The control group of 0 rads ionizing irradiations when actively dividing(Bacq and was placed in the irradiator for the samelength of time Alexander 1461). we can readily see why the nauplii are (40 min) as the group receiving 6000 rads,but in this most sensitive to somatic damage andthe later stages case the safety door was not opened. become more sensitiveto damagein the germinal After irradiation each sampleof 80 animals was tissues. placed in a separate container with Iliter ofculture However, inthefully maturedand reproducing water.Mating was accomplished by allowingthe adults, this sensitivity is masked bythe releaseof one or Artemiato pair and assume the copulatoryposture in two broods of young by each female which apparently the culture water. After the males had clasped the arisefrom oocytes and spermatocytes which have females the pairs were transferredto the individual already been formed from the oogonial and sperma- 500-m1 containers and observed daily to record their togonial cells before irradiation and are awaiting further reproductive performance. This processcontinued until development withinthe organismatthe time of 20 pairs which had received each of thedoses were irradiation. isolated for further study. The first notedpair clasping For the purposes of this experimentit was considered occurred four days after irradiation, and bythe eighth best to irradiatethe animals ata time of maximum day after irradiation all'of the required240 pair matings sensitivity to gamma raysand yet to irradiate them had been isolated. 70

1195

Table 1. Summary of the effects of various dose of irradiation as determined from pair matings of irradiated animals. Based on 20 pairs at each dose. Means and standard errors are given for all per pair and per brood values.

Number Broods Percent Mature adults pose Nauplii voided of per survival Per pair Per brood (fads) broods pair Total Per pair Per brood to adults Total

0 87 4.35± 0.31 9,692 484.6± 30.9 111.4±6.4 64.9 6290 314.5 ± 14.5 72.3±3.9 300 92 4.60± 0.38 9,916 495.8± 29.4 107.8 ± 5.3 67.8 6723 336.2 a 13.0 73.1 ± 3.4 600 89 4.45± 0.39 10,305 515.2± 38.2 115.8 ± 6.8 56.4 5812 290.6 ± 21.3 65.3 ± 5.5 900 60 3.00± 0.30 7,005 350.2± 24.2 116.7 ± 5.2 59.3 4154 207.7±11.1 69.2±5.1 1200 53 2.65± 0.38 8,321 416.0± 34.7 157.0 ± 7.6 50.7 4219 211.0±20.2 79.617.2

1500 39 1.95± 0.28 4,831 241.5± 28.7 123.9± 8.1 40.3 1947 97.4 ± 8.3 49.9 ± 7.2 1800 26 1.30± 0.27 1,894 94.7± 31.2 72.8 ± 8.7 31.1 589 29.4 ± 5.6 22.7±6.4 2100 9 0.45± 0.21 237 11.8± 5.7 26.3±9.2 0 0 0 0 2400 0 0 0 0 0 0 0 0 0

3000 0 0 0 0 0 0 0 0 0 4500 0 0 0 0 0 0 0 0 0 6000 0 0 0 0 0 0 0 0 0

These 20 pair matingsat each of the 12 doses were irradiation on the reproductive performance of the observeddaily during the life span of the female for the brine shrimp,Artemia,when one sex is irradiated and production of broods. If a male was noted as dead in a mated to nonirradiated individuals of the other sex. culture, he was replacedby another male which had Although these experiments are not directly compar- received the same dose.If a female died, that particular able to the work reported here, it would appear that the culture was terminatedat the time of her death. When a results are in accord with the values obtained in this brood was produced, the parents were carefully re- research. moved to afresh culture container and observed daily The number of adults produced per pair as a function for the production of subsequent broods. of dosage is plotted for this experiment in Fig. 2. The The numberof offspring in each brood was counted verticallines included show the magnitude of one on the thirdday after they were produced. They were standard error of the mean above and below the mean. again countedand sexed at maturity. The data obtained The slight increase in reproduction per pair noted at in this experimentare summarized in Table 1. This 300 rads is not statistically significant, but the trend is showsthatthe rangeof doses employed inthe not wholly unexpected. White et al. (1967) have shown experimentcovered therangeof reproductive radio- that brine shrimp which received a dose of 500 rads of sensitivity ofArtemia,since with an acute dose of 2100 gamma rays were longer in size during development and rads, administeredatthe time of maximum radio- matured sexually at an earlier date than control shrimp. sensitivity,the animalsare rendered effectivelysterile, Stimulation of growth by ionizing radiation has also but survivefor essentially as long as the controls. been recognized in various other species, such as crabs A comparisonof this sterility level of 2100 rads for (Engel 1967) and amphibians (Brunst 1965). Such an Anemia withvalues obtained for insects (Grosch and increased growth and early maturity might also result in Erdtnann 1955) would indicate that tlie -reproduction increasedreproduction as possibly indicated inthe performanceof Anemiadisplays a sensitivity to irradia-current experiments. The reality of such an increase will tion which is of the same magnitude as in several orders be examined in the future by the use of inbred strains of insects. Grosch and Sullivan (1954) established a ofArtemiato reduce the within-sample deviations, and sterility dose by use of x rays for Artemia females of the resolution can be increased by using a series of 2250 rads which is also in agreement with the data several dosages between 0 and 600 rads. presented in this experiment. The data from Table l can also be plotted in terms of Recent experiments reported by Squire and Grosch theadults produced per brood versus the dosage (1970) and Squire(1970) detail the effects of gamma administered. Such a graph is shown in Fig. 3. This 71

1196

CONCLUSIONS

Preliminary experiments demonstrated that the stage of the life cycle determined the dosage required to stop reproduction. At the most sensitive stage, the tenth to eleventh instar, a dose of 2100 rads was shown to render the animals effectively sterile. In the study of the various components of reproductive ability it was shown that at doses of 1200 rads or lower,the number of adults produced per brood remained the same. However, the number of adults produced per pair showed a decrease at 900 and 1200 rads, and this decrease can be attributed to a decrease in the number of broods produced per pair. Fig.2. Mean and standard error of the number of adults produced by irradiatedpairs of animals.

LITERATURE CITED

Bacq, Z. M., and P. Alexander. 1961. Fundamentals of radiobiology. 2nd rev. ed. Pergamon Press, New York. 555 pp. Barnes, R. D. 1963. Invertebrate Zoology. Saunders, Philadelphia. 632 pp. Bowen, S. T. 1962. The genetics of Artemia salina. 1. The reproductive cycle. Biol. Bull. 122: 25-32. 1964. The genetics ofArtemia salina.IV. Hybridizationof wildpopulationswith mutant 20 stocks. Biol. Bull. 126: 333-44. Brunst, V. V. 1965. Effects of ionizing radiation on the I I I I I I 1 development of Amphibians. Quart.Rev. Biol. 40: 0 300 600 900 1200 1500 1800 2100 Dose in Rods 1-67. Carpelan, L. H. 1957. Hydrobiology of the Alviso salt Fig. 3. Mean and standard error of the number of adults ponds. Ecology 38: 375-90. produced for each brood of nauplii released. Engel, D. W. 1967. Effect of singleand continuous exposures of gamma radiation on thesurvival and shows rather clearly that the number of mature adults growth of the blue crab,Callinectus sapidus. Rad. produced per brood remains relatively constant until Res. 32: 685-91. the dosage has reached 1200 rads. Then we seea rapid Grosch, D. S. 1962. The survival ofArtemia popula- decrease in the per brood production of adults. Hence, tionsinradioactiveseawater.Biol.Bull.123: some kind of threshold is present at above 1200 rads. 302-16.

Any dosage below this amount has little effect on the . 1965. Biological Effects of Radiation. number of adults produced per brood. Viewing the Blaisdell, New York. 293 pp. same data another way, we can see from Table 1 that . 1966. The reproductivecapacity of the reduction in net reproductive ability demonstrated Artemiasubjected to successive contaminations with at doses of 1200 rads and below, as shown in Fig. 2, is radiophosphorus. Biol. Bull. 131: 261-71. due primarily to a reduction in the number of broods Grosch,D. S., andH.E.Erdman. 1955. X-rays effects produced and not to a decrease in the number of adults on adultArtentia.Biol. Bull. 108: 277-82. produced per brood. At 1500 and 1800 rads wecan see Grosch, D. S., and R. L. Sullivan. 1954. The quantita- a continued reduction in the number of broods pro- tive aspects of permanent and temporary sterility duced as well as a marked decrease in the number of induced in femaleflabrobraconby x-rays and p-radia- adults per brood. lion, Rad. Res. 1: 294=320. 72

1197

Heath, H. 1924. The external development of certain Squire, R. D.1970. The effects of acute gamma Phyllopods. J. Morph. 38: 453-83. irradiation on the brine shrimp,Anemia.II. Female Holton, R. L. 1968. Effects of "Co gamma irradiation reproductive performance. Biol. Bull. 139: 375-85. on thereproductive performance of the brine shrimp, White, J. C., Jr., J. W. Angelovic, D. W. Engel,and E. M. Artenzia.Ph.D. Thesis. Oregon State University. Davis. 1967. Interactions of, radiation, salinity, and Lochhead, J.H.1941.Artenria,the brine shrimp. temperature on estuarine organisms: Effects on brine Turtox News 19: 41-45. shrimp, pp. 33-35 in: Annual Report of the Bureau Squire, R. D., and D. S. Grosch. 1970. The effects of of Commercial Fisheries Radiobiological Laboratory, acute gammairradiation onthebrineshrimp, Beaufort, N.C., for the fiscal year ending June 30, Artenria.1. Life spans and male reproductive perform- 1966. Washington,D.C. (U.S.Fish and Wildlife ance Biol. Bull. 139: 363-74. Service. Circular 270). 73

reprintedfrom:RADIONUCLIDES IN ECOSYSTEMS Proceedings of the ThirdNational Symposium on Radioecology, D. J. Nelson, Editor CONF-710501 (1973)

RLO-1750-89

EFFECT OF GAMMA IRRADIATION ON THE MAINTENANCE OF POPULATION SIZE IN THEBRINESHRIMP,ARTEMIA1 Robert L. Holton2 and CharlesL. Osterberg Division of Biology and Medicine U.S. Atomic Energy Commission, Washington, D.C. William O.Forster ` Puerto Rico Nuclear Center College Station, Mayaguez, Puerto Rico

ABSTRACT

Population cultures of the brine shrimp, Anemia, were irradiated with acute doses of gamma irradiation from a 60Co source. The subsequent reproductive performance and size ofthe cultures were studied for a period of 20 weeks. It was demonstrated that the population cultures may be maintained with only a small part of the reproductive potential exhibited in the pair matings. Therefore, we find that the resultsof pair matings must necessarily be used to assess the amount that the reproductive potential ofArtemia is decreased due to various doses of irradiation. Population cultures at all doses were shown to have the same sized populations at the end of 20 weeks. However, it was demonstrated that the populations irradiated at higher doses had not recovered to their full reproductive potential at the end of this time.

INTRODUCTION required touse 1% ormore of theirpotential to maintain thesamepopulationsize. Hence, although we Although it is possible to examine the radiosensitivity may not detect the effects of radioactivecontamination of the brine shrimp under the most sensitive conditions when studying laboratory populationcultures, such (Holton et al., this symposium), such a study does not effects could be manifest in a competitive environment. effectively assess the effect of irradiation on a popula- Of particular interest, however, are Grosch's (1966) tion composed of all age classes. The work described results which show that the addition of as little as 20 here is an attempt to assess the impact of an acute dose pCi of 6 s Zn to a 3-liter population culture ofArtemia of gamma radiation on the subsequent reproductive has consistently caused the cultures to become extinct. performance of population under rather carefully con- In the case of 32P, the populations are able tostand the trolled laboratory conditions. addition of greater amounts of activity. Cultures rou- Grosch (1962_, 1966) has demonstrated the effects of tinely survive the addition of an aliquot of 30 yCi of the addition of 65Zn and 32P to 3-liter population 32p per 3-liter culture, and oneculture has survived the cultures ofArtemia.He has concluded that although addition of as much as 90 pCi of "P.P. the number of adults counted in population cultures At least a part of the explanation for this different may be the same, the animals from experimental response to the two nuclides may be in terms of the cultures haveashortenedlifespan, deposit fewer fact that the 65Zn has a 245-day physicalhalf-life and zygotes per brood, and show poor survival to adulthood the 32P has only a 14-day half-life. A culture receiving as compared with control animals which have not been the addition of 20 pCi of 6 5 Zn wouldbe exposed to a subjected to radionuclide contamination. He also has higher level of activity over a longer period of time than determined that population cultures of control animals a culture receiving 20 pCi of 32P. Therefore, such a use only 0.2% of the reproductive potential that they 65Zn culture would be exposed, during the course of exhibit in pair matings to maintain the population size. the experiment, to many more radioactive decay events However, populations of experimental animals are than a 32P culture which receiveda higher initial level of activity. 1. Information contained in this article was developed during However, in this study we are interestedin deter- the course of work under fellowship number 1-F1-WP-30, 166-01 of the USPHS and 1-Fl-WP-30, 166-02 of the Federal mining what part the ionizationin livingtissue plays in Water Pollution Control Administration. the observed effects, as compared with the effects due 2. Present address: Department of Oceanography, Oregon to transmutation and also due to the recoil energy of State College, Corvallis, Oregon 97331. the decaying nucleus. The present study,using an

1198 74

1199

external source of gamma rays from "Co, will allow for later countingwere includedin the original irra- for some degree of assessment of the effects of certain diated sample. selected doses of ionizing radiation on population The Artemiairradiated forthis experiment were maintenance of the brine shrimp, without having to hydrated,hatched,and handled by thestandard consider transmutation and recoil effects as potential methods. A series of ten different groups of eggs were sources of difficulty in the interpreting of the results. hydrated every other day, and as they hatched, they In particular, this series of experiments was designed were introduced into an 8-gal aquarium for continued toestablishdosagelevels,inrads,at which the development. The aquarium stock was fed daily in reproductive ability ofArtemiawas affected and to amounts necessary to maintain a normal growth rate. determinethe dose required to reduce the reproductive When the oldest animals in the aquarium had matured ability of the population to zero and thus fix the level and were starting to produce young, samples ofArtemia at which a laboratory population would be unable to were prepared for irradiation. reproduceitself and hence go to extinction under A random sample of animals was removed from the laboratory conditions. aquarium by use of a dip net, after the animals had been thoroughly mixed to ensure a sampling of all sizes METHODS ANDMATERIALS of animals. The animals removed were placed on an I100-p bolting silk. A double rinse with fresh culture The organism used in this work and its life cycle have water of the animals on this filter allowed the few early been described previously (Holton 1968, Holton et al., stages present to pass through and retained essentially this symposium). The particularArtemiaused for this all animals from the seventh instar to the adults on the work were derived from the population in the Great bolting silk. The animals retained on the bolting silk Salt Lake in Utah. were then divided into groups of 300 animals. This size The pairmating irradiation experiment (Holton et al., of culture was chosen in view of the report of Grosch this symposium), which required the irradiation of (1962) that as many as 300 well-developedArtemia animals ata specific stage of their life cycle, allowed were counted in his 3-liter population cultures at one more accurate control of experimental conditions than time. was possible in the experiment to be described in this Four groups, of 300 animals each, were then selected section. However, such irradiation fails to simulate the at random for irradiation by the standard technique at conditions of exposure to irradiation that a functioning each of the following dosages: 0, 500, 1500, and 3000 population might encounter in an ecological setting. In rads. The four control groups were placed in the the environment, a typical animal population often irradiatorfor the same length of time as for the consists of all stagesin the life cyclebeing represented 3000-radgroups, butthe safety door was not opened. at any given time. The population culture irradiation Each group of 300 irradiated animals was then placed in experiment described in this section is an attempt to 3 liters of standard culture water in a gallon jar to start approach the environmental situation by irradiating a a separate populationculture. All of thesepopulation population composed of animals from the seventh cultures were fed 1 ml of the standard yeast suspension instar describedby Heath (1924) to mature adults daily. which were reproducing at the time of irradiation. Irradiation of the earliernaupliistages was not RESULTS AND DISCUSSION included in this experiment for two reasons. First, at the .levelsof irradiation to be used, the subsequent The effects of this irradiation upon the population reproductive performance of these early nauplii would maintenance were studiedin two different ways. Three be confoundedwith their altered developmental rate. A of the cultures at each dosage were examined every four second reason for irradiating only the large nauplii and weeks for a 20-week period in the following manner. adults wasthe impossibility of mechanicallyseparating The entire culture was siphoned out of its container and the smallerstages from the solid waste materials which carefully strained through an 1100-p bolting silk. The accumulate in the population cultures. Because of this animals on the bolting silk were rinsed twice with water problem, it was possible to count only individuals from decanted off the top of the culture to ensure removal of the seventh instar to the adult state during the later all young nauplii. This procedure allowed all of the counts of the population culture. Hence, to ensure a smaller nauplii to pass through the bolting silk and uniformity in determining the numbers of animals in againretained the animals more mature than the the cultures, only those stages which could be recovered seventh instar. A population count of these stages was 75

1200

Table1. Mean number of seventh instar to adult Artemia found in populationcultures receiving various doses of irradiation. Each value is the mean and standarderror, basedon threereplicate population cultures.

Sampling Dose (rads) interval (weeks) 0 500 1500 3000

0 300± 0 300 t 0 300± 0 300'± 0 4 578± 29 426331 337± 27 284x27 8 598± 39 603 ± 43 341± 15 354 ± 24 12 513± 24 627 ± 20 399± 25 385 ± 16 16 453± 25 520 ± 14 421± 18 344 ± 25 20 375± 36 379 ± 35 362± 20 337 ± 12

then made. After counting was completed, the entire 4 8 12 16 20 population was returned to the culture container with WEEKS AFTER IRRADIATION original culture water for further development. The Fig.1. Mean number of seventh instar to adult Artentie data obtained from these counts are presented in Table found in population cultures receiving various doses of irradia- 1. tion. The mean value calculated for each dosage at the four-week intervals is plotted in Fig.1. Several things not possible to explain this lag. It was shown that at the should be noted about this evaluation of the irradiation most radiosensitivestageof the life cycle,a dose of 600 effects by total population counts. The number of rads had no significant effect on the numberof adults animals found in the control populations had virtually produced perpair,as compared with the control doubled at the end of four weeks, indicatinga real animals, and yet in this experimentwe see a marked potential for population growth under these culture effectatthe end of the fourthweek between the conditions. However, a comparison with the control control animals and those which received 500 rads. animals in pair mating experiments (Holton et al., this The similar decrease in population size noted for both symposium) shows that only about 1/80 of therepro- the 0- and 500-rad cultures in the finalweeks of the ductive potential inherent in this population ofArtemia experiment and the fact that the culturesreached was realized under these population culture conditions. almost identical mean values on the 20th week after This fact leads us to conclude that an environmental irradiation may be interpreted to meanthat there was stress was severely limiting the population size. This somefinitecarryingcapacity inherentunder these conclusion was supported by the observation that the culture conditions, and that all thepopulations were control populations increased only slightly during the tending to approach this carrying capacity asa limit. next four weeks, and during the final weeks of the Although a population, with little or nodamage from experiment the number of animals inthe control irradiation, may temporarily overshootthis carrying cultures steadily decreased. Figure 1 shows that this capacity,theenvironmentalstresswillultimately decrease was linear over the final 12 weeks of the mount and reduce the population toa size com- experiment and had shown no signs of leveling off at mensurate with the potentialities of the environment. the termination of the experiment. It may be assumed The cultures receiving the doses of 1500 and 3000 that this control population would have reached an rads showed the effects of the irradiationduring the even lower level if the experiment had been continued early weeks of the experiment with decidedlysmaller for a longer period of time: populations than the control cultures; hence for the The threeculturesreceiving 500 rads showed a first 12 to 16 weeks the radiation exposurelimited the pattern of distinct population growth followed by a size of these populations. However, at the end of 20 steady decline which was very similar to the control weeks all of the populations appeared to be approach- culture. However, in this case a four-week lag in the ing a common size, which apparently was controlled by growth pattern, as compared with the control popula- environmental conditions discussed above. This might tions, was observed. In view of the data presented in also be used as evidence that since after 20 weeks or previous work (Holton et al., this symposium) it was about five generations, even the population which 76

1201

received 3000 rads was maintaining itself about as well as the control population, and therefore had essentially recovered from any reproductive damage resulting from the insultwith gamma irradiation. However, this hypothesis of little or no reproductive damage being evident after five generations was not supported by the O results of the second part of this experiment cited pR below. It was shown that the individual pairs from the m0. population _QZ cultures receiving 1500 and 3000 rads tti showed a significantly lowered reproductive potential whencomparedwith the control cultures. Hence, a more realistic assessmentof the situation would be to t conclude that after 20 weeks the populations of all four cultures were beinglimited by environmental factors to

I essentiallythe same size. W I i 1 I I 0 4 8 12 16 20 The fourth culture receiving each of the four different WEEKS AFTER IRRADIATION doses was handled in a different way. In this case, ten individualpair matings of breeding adults were isolated Fig. 2. Mean and standard error of the number of broods produced per pair of Artemia. removed from irradiated popula- from each culture at four-week intervals. The mated tions at various time intervals. pairs of brineshrimp were selected randomly from the cultures. As a pair appeared at the top of the culture, theywereremoved with a small dip net and placed in a 500-m1containerwhere their reproductive performance specificfactscanbe seeninFig.2. The brood was studiedfor the life span of the female, with the production was affectedrelativelylittleat the first sampling period, which occurred within 6 hr of the time same techniqueusedinapair matingexperiment (Holton etal.,this symposium). Such a method of the animals were irradiated. At this time many females selectingpairs ofArteiniafor study chose only those already had one or two broods of eggs which were pairsat the surface of the culture, and if such pairs were maturing in their oviducts or uteri. The fertilization and not a random sample of all pairs in the culture, this release of these eggs provided the relatively less affected procedure was biased against those pairs which spend a brood production at this first sampling interval. majority of their time at the bottom of the culture. At the four-week sampling period we saw a significant This procedure would be especially biased if the pairs decrease in brood production for all irradiated samples. swam at thesurface of the culture only at a certain age The decrease for the 500-rad group stands in contrast to and thereforeeliminatedother age classes from the the 600-rad group in a pair mating experiment (Holton sample. et al., this symposium) where no significant decrease in The data obtained from the study of thepair matings brood production was noted. The only explanation that made fromthe population cultures are summarized in can be advanced to explain this phenomenon would be Table 2. However, itis much more informative to that the effects on brood production will tend to show extract specific values from Table 2 and to plot these up to a greater degree within the first filial generation values to study various individual components of the that was sampled at four weeks in this experiment than reproductive performance ofArtemiaand, therefore, to in the generation which was exposed to the irradiation. assess therelative importance of the various parameters The importance of this possibility would suggest that it in the reproductive performance of the species. warrants intensive study in the future. Figure 2 shows the mean number of broods produced Theprincipalinterestinthis population culture by each pair removed from the cultures for study irradiation experiment was to ascertain the degree of during the course of the experiment. Since it was shown recovery of reproductive performance at successive in a pairmating experiment (Holton etal.,this intervals after irradiation. Figure 2 showed that the symposium)that brood production was the most brood production after the fourth week began a slow sensitive parameter whenArtemiawere insulted by but constant recovery. We concluded that at doses of irradiation, it was interesting to note that in this study 1500 rads orover, more than 20 weeks, or five the brood production was also shown to be greatly generations, are required for the brood production to affected by the doses of gamma rays involved. Several return to the control level. 77

1202

Table 2. Summary of the effects of various doses of irradiation on population cultures of Anemia, as determined by pair matings from the population cultures. Based on ten pairs at each sampling interval. Means and standard errors are given for all per pair and per brood values. Sampling Number Broods Nauplii Mature adults interval of per survival - (weeks) broods pair Total Per pair Per brood to adults Total Per pair Per bro,ud

3000 rads

0 26 2.6 ± 0.44 1821 182.1 ± 31.2 70.0± 11.8 20.9 381 38.1± 20.4 14.7 12 4 4 22.9 154 15.4 6 0.6 ± 0.37 671 67.1±28.0 111.8± 16.8 ± 12.0 25.6 14 8 7 0.7 ± 038 664 66.4 ± 27.9 94.9± 15.2 26.7 177 17.7± 12.3 25.3 la 12 11 1.1 ± 0.40 837 83.7±28.5 76.1± 10.1 27.4 229 22.9± 20.6 20.8 14, 16 9 0.9 ± 0.41 796 79.6 ± 27.0 88.4± 11.0 33.5 267 26.7± 19.8 29.7 lu s 20 16 1.6 ± 0.53 1078 107.8 ± 28.8 67.4± 12.3 37.7 406 40.6± 21.1 25.4 13i 1500 rads 0 29 2.9 ± 0.52 2983 298.3 ± 35.6 102.9 41.8 124.7 ± 11.8 1247 1 30.5 43.11 if 1 4 16 1.6x0.45 1516 151.6 ± 29.4 94.8± 13.0 33.3 505 50.5± 20.7 31.6.6!1 8 20 2.0 ± 0.68 1873 187.3 ± 30.9 93.6± 11.4 34.7 650 65.0± 17.7 32.5 8 9 12 23 2.3 ± 0.47 2008 200.8 ± 31.0 87.3± 12.7 44.6 896 89.6± 16.0 39.0 7 4 16 24 2.4 ± 0.52 2367 236.7 ± 30.3 98.6± 12.5 41.8 989 98.9± 14.1 41.2 t 9 0 20 29 2.9 ± 0.40 2739 273.9 ± 33.5 94.5± 12.3 51.0 1397 139.7± 18.9 48.2 7.5 500 raids 0 40 4.0 ± 0.63 4444 444.4 ± 39.7 111.1± 11.7 55.6 2471 247.1± 24.5 61.8 ± 6.7 4 27 2.7 ± 0.53 3789 378.9 ± 38.2 140.3± 11.9 53.2 2016 201.6± 20.2 74.7 6.1 8 33 3.3 ± 0.66 3435 343.5 ± 36.7 104.1± 9.3 54.4 1869 186.9± 23.8 56.6 ± 8 1 12 31 3.1 ± 0.37 3678 367.8 ± 28.1 118.6*_ 10.4 56.6 2082 208.2± 22.2 67.2 t 7 3 16 36 3.6 ± 0.48 3798 379.8 ± 37.8 105.5± 9.8 54.7 2078 207.8± 27.1 57.7 ± 6.0 20 39 3.9 ± 0.70 4259 425.9 ± 42.9 109.2± 10.5 59.0 2513 251.3± 21.9 64.4 ± 5 7

0 rads 0 42 4.2 ± 0.64 4636 463.6±42.6 110.4± 13.0 62.1 2879 287.9± 27.3 68.5 ± 7.8 4 39 3.9 ± 0.56 4121 412.1 ± 37.8 105.7 ± 12.7 70.1 2889 288.9± 24.0 74.1 s 6.7 8 38 3.8 ± 0.41 4227 422.7 ± 38.9 111.2 ± 12.4 65.6 2773 277.3± 23.5 73.0 t 7.2 12 43 4.3 ± 0.65 4239 423.9±41.1 98.6±11.0 60.1 2548 254.8± 26.4 59.3 t 7.1 16 38 3.8 ± 0.61 3712 371.2±39.3 97.7±11.2 67.4 2509 250.9± 21.6 66.0 ± 6.8 20 42 4.2 ± 0.55 4133 413.3 ± 40.1 98.4 ± 10.9 62.0 2562 256.2± 24.1 61.0 ± 8.5

A second parameter involved in the reproductive decrease in nauplii production and also showed again performance is plotted in Fig. 3. Here we see that the that after 20 weeks the populationhad not recovered number of nauplii produced on successive weeks by from the reproductive damage dueto irradiation. various samplesof females from irradiated populations A thirdfactor contributing to total reproductive shows essentially the same trendas seenin brood performance is revealed by Fig. 4. Here thepercentage production. However, we do note some differences. The of nauplii produced which surviveto adulthood is animals from the 3000-rad culture showed no signifi- plotted. The control populationis shown to rather cant increase in nauplii production from the 4th to the consistently have a 65% survival of nauplii. This value 20th week, even though the number of broods pro- was higher than average control survival reportedb duced was shown to increase. Hence, our attention is Grosch in 1962, but would be well within the range of called to Table 2, where we confirm this trend by survival values that Grosch (1966) reported in later noting a decrease in the number of nauplii produced per experimental work. brood from the 4th to the 20th week. No explanation is The percent surviving for each of theexperimental apparent to explain such a trend. The difference in groups showed ageneral tendencyto increaseat number of nauplii produced between the control and successive sampling periods withthe population irra- the 500-rad culture was shown to be slightat all times. diatedat 3000 rads, showing a steadyincrease in However, the 1500-rad culture showeda significant survival. At 1500 rads, we see thesame trendof 78

1203

determining the survival rate at the different times. At the first sampling period, survival was a function, at least in part, of the ability of zygotes already formed to withstand the effects of irradiation. Figure 4 shows that at this sampling period, 3000 rads were responsible for a pronounced decrease in survival of such zygotes when compared with the survival at 1500 rads. However, at the end of four weeks the survivalrate should be interpreted as a measure of the amount of genetic damage carried in the germinal tissue, and we see that at this time the difference between the effects on survival of 3000 and 1500 rads was much smaller. Another parameter is presented in Fig. 5. Here, the number of mature adults produced per brood was the same for both the control and the 500-rad populations. WEEKS AFTER IRRADIATION Only at the 1500- and 3000-rad doses do we find the number of adults produced per brood decreased due to Fig. 3. Mean and standard error of the number of nauplii the effects of the irradiation. produced per pair of Anemia, removed fromirradiated popula- It should also be noted that at four and eight weeks tions at various time intervals. the difference in number of adults produced per brood was very slight for the 1500- and 3000-rad populations. However, from the12th week to the end of the experiment, we noted that the 1500-rad culture made a marked improvement in the number of adults produced per brood, while the 3000-rad population showed little °'_500 r improvement. 50 The final parameter is really the factor of ultimate concern and depends upon each of the previously discussed components of the reproductive ability. This net reproductive potential within the laboratory conditions of this experiment is presented in Fig. 6 as the total number of sexually mature adults which were

1 I r 20'r 1 1 0 4 8 12 16 20 WEEKS AFTER IRRADIATION Fig. 4. Percent of nauplii, produced by parents removed from irradiated populations at various time intervals, surviving to adulthood.

improved survival from the 4th week to the 20th week. However,itisinterestingtonotethe difference between the 3000- and the 1500-rad populations at the zero- and four-week sampling intervals. The 1500-rad population showed a very definite decrease in survival between these two periods. However, with the 3000-rad 4 12 16 20 sample we see that the lowest survival rate occurred at 0 8 WEEKS AFTER IRRADIATION the first sampling period and that at four weeksa slight improvement was already noted. This differential pat- Fig. S. Mean and standard error of the number of adults tern of survival at the two doses may be interpreted to produced per brood by Arternia, removed from irradiated indicate that different mechanisms were operating in populations at various time intervals. 79

1204

The results presentedinFig. 6 show thata real decrease in reproductive potential was experienced in both the 1500- and 3000-rad cultures and was still present at the end of 20 weeks. The total population counts indicated that the 1500- and 3000-rad cultures were maintaining their population site as well as the control culture by the end of 20 weeks. However, we see here that though they were maintaining a normal population site, they still show the effects of irradiation by the display of this lowered reproductive potential.

CONCLUSIONS

The observed effect of a particular dose of irradiation depends on the nature of the observation. Counts made of population cultures which received different doses I I 1 1 1 may show the same population levels and hence 0 4 8 12 16 20 indicate that those populations which have received the WEEKS AFTER IRRADIATION higher doses have recovered from the effects of irradia- Fig. 6. Mean and standard error of the number of adults tion. Pair matings from these same cultures have been produced per pair of Artemia, removed from irradiated popula- shown to indicate very different reproductiveabilities tions at various time intervals. between the populations which have received different doses. This fact should be kept in mind, especially when thinking about the effects of irradiation in the environ- produced by each pair of Artemia from the irradiated mental situation. Even if populationcultures, under populations. ideal laboratory conditions, show no effects of tlffi We note that the pair matings isolated from the irradiation at a particular dose, it is possible that their control culture, on the average, produce about 270 reproductive potential has been decreased. This same sexually mature adults per pair. The lower production dosage might affect the functioning field population exhibited during the final three sampling periods as which is subjected to competition with other species as compared with the first three sampling intervals cannot wellasenvironmentalstress. Under theseenviron- be readily explained. The population culture exposed to mental conditions, a population could conceivably be 500 rads suffered an initial significant depression in net driven to extinction with a radiation dose that showed reproductive rate, but the results indicated that it had little or no effect in a laboratory populationculture. approached complete recovery at the end of 20 weeks. The results presented by the analysis of pairmatings In the case of the culture which was exposed to 1500 in the population culture experiment indicated that rads, we see that the net reproductive rate dropped to after a relatively long time of 20 weeks or about five less than 20% of the control rate four weeks after generations, the populations which received doses of irradiation. It then began a progressive recovery and, by 1500 and 3000 rads were stillfar fromrecovering the final week of the experiment, exhibited about 50% completely from the effects of irradiation. This is in of the reproductive potential of the control culture. agreement with the results of Grosch (1962, 1966), The production of mature adults by pair matings who concluded that a minimum of 12 generations must from the culture receiving 3000 rads dropped to less elapse before a population culture which has survived than 10% of the control culture. Although this culture the addition of 30 pCi of 32P per 3-literculture can exhibited an improvement in reproductive performance recover to the point of surviving the addition of a over the course of the experiment, at the final sampling second dose of 30 pCi of 32P. the reproduction was still only about 20% of the rate Further study is suggested in certain areas by the exhibited by the controls. Itis important to point out results presented here. The results discussedearlier, that the rate of recovery of net reproductive ability was indicating that the effects of irradiation on the number greater for the 1500-tad population than it was for the of broods produced may be greater with the first filial 3000-rad population. generation than with the generation that was exposed 80

1205

toirradiation, are important enough to warrant inten- reproductivepotential.During the course of such sive study in the future. The levels for the production recovery, the nature of the existing ecosystem might be of sterility and for affecting various parameters of total irreversibly altered. reproductive performance can be determined more preciselyby using a series of more closely spaced doses LITERATURE CITED to refinethe resolution of the present experiments. It Artemia wouldalso seem necessary to extend the present Grosch, D. S. 1962. The survival of popula- experiment by the use of a low-level source to provide tionsinradioactiveseawater.Biol.Bull.123: for chronic irradiation of theArtemiapopulation. Such 302-16. experimentswould allow interesting comparisons, both 1966. The reproductive capacity of with this research and with the results of Grosch Artemiasubjected to successive contaminations with (1966). radiophosphorus. Biol. Bull. 131: 261-71. Inconclusion, we should notethat the results Heath, H. 1924. The external development of certain presentedhere would tend to confirm the theory that Phyllopods. J. Morph. 38: 453-83. irradiationatthelevels presently occurring in the Holton, R. L., W. 0. Forster, and C. L. Osterberg. The environmentwill have little or no detrimental effect on effectof gamma irradiation on the reproductive the reproductive ability of animal populations. How- performance ofArtemiaas determined by individual ever, the results do indicate that in the event of greatly pair matings (this symposium). increasedenvironmental contamination, any reproduc- Holton, R. L. 1968. Effects of "Co gamma irradiation tive damage sustainedwill persist over a long period of on the reproductive performance of the brine shrimp, timebefore thepopulationscan recover their full Artemia.Ph.D. Thesis. Oregon State University. 81

RLO-2227-T12-1

Calif. Mar. Res. Comm., Ca1COFI Rept., 16:66-73,1972

ZINC-65 AND DDTRESIDUES INALBACORE TUNA OFF OREGON IN 1969

WILLIAM G. PEARCY Department of Oceanography Oregon State University Corvallis, Oregon and

ROBERT R. CLAEYS Department of Agricultural Chemistry Oregon State University Corvallis, Oregon

THE HYPOTHESIS to the ocean. Zinc-65 and chromium-51 (half lives of 245 and 28 days respectively) were the most common Albacore(Thunnus alalunga) istypicallyan artifically induced radionuclides entering the Pacific oceanic tuna found far from continents. However, Ocean (Osterberg, Pattullo and Pearcy, 1964; Gross, each summer a portion of the North Pacific popula- Barnes and Riel, 1965; Perkins, Nelson and Haushild, tion migrates into nearshore waters along the west 1966; Osterberg, 1965). coast of North America. Pearcy and Osterberg (1968) The distribution of Columbia River water in the found that the zinc-65 radioactivity increased dra- ocean varies with season and is dependent upon the matically in albacore caught off Oregon and Wash- direction of prevailing winds. During the summer, ington during the summers of 1962-1966. This in- northerly winds predominate and Columbia River crease was attributed to the association of albacore waters usually flow to the south or southwest as a with Columbia River water, a known source of radio- "plume." The maximum discharge of the Columbia nuclides into these marine waters. River usually occurs in June, and the plume can often Manigold and Schulze (1969) reported DDT, DDE be detected by low salinities far to the south of the and DDD in Columbia River water during various mouth of the river (Barnes and Gross, 1966; Owen, seasons, 1966-1968. Sometimes levels of residues were 1968). With the onset of winter, the wind stress 0.1 ppb (0.1 µg/1), which is similar to those reported reverses and prevails from the south, resulting in in the Sacramento River which drains agricultural northerly transport of Columbia River waters along lands where pesticides are heavily used. the Washington coast. During this season the plume Therefore, if the Columbia River is a major mech- is not as obvious because of large runoff of coastal anism for transport of DDT and its metabolites to the rivers (Barnes and Gross, 1966). Figures 1 and 2 ocean, then DDT as well as 65Zn should increase with illustrate the location of plume waters during the time while albacore inhabit nearshore waters off summer by salinity and the winter by chromium-51. Oregon during the summer. Similar trophic pathways The plume has been identified by chromium-51 and and uptake rates for DDT and 65Zn are implicit in zinc-65(Gross, Barnes and Riel, 1965; Osterberg, this hypothesis. Cutshall and Cronin, 1965; Barnes and Gross, 1966), specific alkalinity (Park, 1966) and surface temperature (Owen, 1968; Pearcy and Mueller, 1970; Pearcy, BACKGROUND in press). The Columbia River is the largest river on the Seasonal, geographic and bathymetric variations in Pacific coast of the Americas and the second largest the Zn-65 content of marine organisms off Oregon in the United States. It discharges about 7,250 m3 is correlated with the distribution of Columbia River per second of fresh water into the Pacific Ocean at waters at sea (Carey, Pearcy and Osterberg, 1966). the headwaters of the California Current (Leopold, Pearcy and Osterberg (1967) found seasonal maxima 1962). The Columbia is also unique because until 1971 during the summer in Zn-65 of oceanic zooplankton it carried distinctive radionuclides from the Ilanford and micronekton collected by midwater trawling in nuclear reactors into the ocean.' Columbia River the upper 150 in off Oregon. These seasonal variations water was used as a primary coolant for the reactors (see Figure 3) are believed to result from seasonal and the intense neutron flux of the reactors induced shifts in the location of plume. radioactivity of trace elements in the water. Stable Zinc-65 was ubiquitous in marine organisms radiozinc-64, for example, added a neutron and became analyzed off Oregon (Osterberg, Pearcy and Curl, radioactive zinc-65. Most of the radionuclides in the 1964; Carey, Pearcy and Osterberg, 1966). Of particu- coolant water had short physical half-lives and de- lar interest is the enhancement of Zn-65 in albacore cayed to undetectable levels during the 370 mile trip tuna. Zinc-65 levels in albacore livers increased dra- 'When this paper was written (February 1971) all nuclear reac- matically during the summer months off Oregon and tors atHanford were shut down. Washington, whereas levels in albacore collected off (66) 82

REPORTSVOLUME XVI, 1JULY 1970 TO 30 JUNE 1971 67

SURFACE SALINITY COLUMBIA RIVER CRUISE COOC 4 a 5 SICr SURFACE ACTIVITY 18 JUNE -3 JULY 1969 (PC/liter)

FIGURE 1. Pattern of Columbia River plume based on surface salinity, June 18-July 3, 1969. FIGURE 2.Pattern of Columbia River plume based on chromium-51 southern and Baja California were much lower and activity of surface water, February 1966 (from Frederick, 1967). displayed no seasonalincrease(Figure4).This marked increase in Zn-65 levels off Oregon and Wash- and samples of livers and flesh obtained in the labora- ington is attributable to albacore feeding on animals tory ashore. These samples were used for radioanalysis which have accumulated Zn-65 introduced into the and pesticide analysis. ocean by the Columbia River. Albacore concentrations Procedures for radioanalysis were similar to those are also known to be located in the vicinity of plume described by Pearcy and Osterberg (1968). Whole waters, probably because plume waters are heated livers and samples of flesh were weighed, dried, ashed more rapidly than surrounding waters during the and packed in 15 cc counting tubes for radioanalysis early summer(Owen,1968;Pearcy,in press; Pan- in a 12.7 cm3 sodium iodide (TI) crystal coupled to shin, unpublished). a 512 channel pulse-height analyzer. Counting time Because of the marked increase in Zn-65 in albacore was 100 or 400 minutes. after they migrated into Oregon waters and because The predominant gamma emitter was zinc-65. To of the DDT residues detected in Columbia River water permit calculation of specific activity of zinc-65, total (Manigold and Schulze,1969),we examined albacore concentration of zinc was determined by atomic ab- in 1969 for both radionuclides and pesticides. sorption spectrometry (Perkin-Elmer model 303) on samples of ash digested in nitric and hydrochloric METHODS acids. Liver samples for pesticide analysis were obtained Albacore (550-730 mm forklength)were collected from lateral lobes, the median lobe, or entire homoge- off the Oregon coastduring thesummer of 1969 oil nized livers. Flesh samples were taken near the body jigs trolled from theR/V YAQUINA and CAYUSE cavity. The tissue sample was diced and a 7-20 grain or the fishboatsSUNRISE and TYI'IIOON.Livers subsample was blended in an Omni-mixer with 25 ml were either removed from the albacore at sea and of acetone and 1 gram of anhydrous Na2SO4 per gram frozen in plastic bags, or whole albacore were frozen of sample. After filtering the mixture through a Buch- 83

68 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS

b N a

- 500-1000 m

_! ` 0 1 1 I L I I-T- I - II A M J J A S 0 N D J F M A M J J A S 0 N 0 J F M 1963 1964 1965

FIGURE 3.Zinc-65 radioactivity per gram wet weight from opening-closing midwater trawl samples from three depth intervals, 93 km off central Oregon coast, 1963-1965. Solid lines connect the average of nighttime collections, the dashed lines of daytime collections. Vertical lines show the range of values (from Pearcy and Osterberg, 1967). NORTHERN OREGON-WASHINGTON SOUTHERN OREGON SOUTHERN 8 BAJA CALIFORNIA (45 - 305 *m) (4/5- 555 km ) (/520 - /900 km) I

1000 r- I- A

o 100

10 1 E JULY AUG. SEPT. OCT. JULY AUG. SEPT. OCT. JULY AUG. SEPT. OCT. FIGURE 4.Zinc-65 per gram of albacore liver ash off the west coast of North America during the summer of 1962 (0), T1963 (-F), 1964 (S), 1965 ( A Oregon, A Washington), and 1966 (0). The distances in kilometers indicate how for albacore captured in the three regions were from the mouth of the Columbia River (from Pearcy and Osterburg, 1968). 84

REPORTS VOLUME XVI, 1 JULY 1970 TO 30 JUNF.1071 69

ner Funnel, the vacuum was reduced, the filter cake microcoulometric detectors with a sensitivity (0.1 my covered with acetone, and vacuum was reapplied. One peakheight)of 0.01 and 1.0 nanograms of p, p'-DDE part water per part acetone was added to the filtrate respectively.A 122 cm (4 ft) x 2 mmi.d. pyrex and pesticides were extracted with 50 ml of pentane column (fluorosiliconepolymer)was packed with a in a separatory funnel. The pentane layer was col- 2.4:1.0 mixed packing of 7% QF-1 and 7% DC-11 lected and an additional one part water and 50 ml of (methylsiliconepolymer)on 100/120 mesh Chromo- pentane were added for a second extraction. The com- sorb W high performance for analysis. bined pentane fractions were evaporated to 5-10 ml The conditions for the electron capture gas chro- and loaded on top of a Florisil adsorption column for matographwere:210°C inlet temperature, 180°C further separation. columntemperature,200°C detector temperature and The Florisil column was prepared by adding 25 ml column gas(nitrogen)flow rate of 40 ml/min. The of hexane to a 16 mm i. d. glass column fitted with a: retention time of p, p'-DDE was approximately 4 Teflon stopcock. Ten grams of -200 mesh, 450°C minutes.The sample solution was analyzed in dupli- oven dried Florisil were stirred. into the column and cate by injecting 2-8 µl subsamples into the gas chro- washed down with hexane. Then I inch of anhydrous matograph with a 10 µlsyringe,followed by injections Na2SO4 was added. of 2-8 µl subsamples of 0.01, 0.02 or 0.04 ng/µl stand- Pesticides were eluted from the column with: 1st ardsolutions.The amount of standard was selected cut, 60 in] of hexane; 2nd cut, 75 ml of 5% benzene to give a peak within 20: of the sample pesticide hexane (this eluant contains p, p'-DDE, o, p'-DDT, peak.The second injection of sample solution was p, p'-DDT) ; 3rd cut, 75 ml of 20% benzene-hexane either } or twice as large as the firstinjection.A third (contains p, p'-TDE) ; 4th cut, 60 ml of 15% diethyl analysis was required if the peak heights were not ether, 0.25% acetone, 85% hexane; 5t)1 cut, 90 ml of within 8% on a comparable basis. the 15% diethyl ether, 0.25% acetone, 85% hexane Because of the very low levels of pesticides present, mixture (contains dieldrin and endrin). After initial special precautions were taken to avoid contamination. analysis showed that DDE could elute into the first Technical acetone was refluxed with KMnO4(0.5 g/1) cut on occasion, the hexane elution was omitted. The for 1 hour anddistilled.N-pentane and n-hexane were 4th and 5th cuts were not analyzed routinely because refluxed over sodium wire overnight before distillation. of poor lipid separation and absence of pesticides. The Florisil adsorbent required purification and was The column eluant was analyzed for pesticides by placed in a muffle furnace at 450°C for 15 hours. All gas-liquid chromatography using electron capture and glassware was isolated and cleaned with K2CrO4-

1000 0.10 I I

0 0 0 OCA 0

8D iD 0 00 Co 0 0.01 0 0 It V 0

IO 0.001 JUNE JULY I AUGUST SEPTEMBER FIGURE 5.Zinc-65 radioactivity (0) and specific activity, ssZn/Zn (o) of ashed samples of albacore livers, 1969. 85

70 CALIFORNIA COOPERATIVE OCEANIC FISHERIES INVESTIGATIONS

H2SO4 cleaning mixture and rinsed with water and all metabolites were low. p, p'-DDE was the most acetone. The anhydrous Na2SO4 was washed with prominent residue. It was found in concentrations up hexane before use and the water was purified by hex- to it maximum of 0.16 ppm and was present above ane extraction prior to use. The contamination of the blank values in all 41 samples analyzed. Chroma- glassware and reagents was determined prior to use tographic peaks attributable to p. p'-DDT ;o, p'-DDT by analyzing ''solvent blanks",i.e,following the and p, p'-TDE were usually present. Although levels above procedure without addition of tissue. of these residues were lower than p, p'-DDE, most Our procedure for the elution of the Florisil column values were significantly above blank values. Low was selected for analysis of low levels of chlorinated values of o, p'-DDT and p, p'-DDT may be influenced hydrocarbons. Other reported procedures resulted in by PCB's. No dieldrin was detected above blank levels interferences in talc ,its chromatographic analysis. of 0.003 ppm. The average values for all DDT resi- The presence of DDT was confirmed by microcoulo- dues combined were 0.05 for liver (range 0.003--0.22) metric detection of halide compounds and by base and 0.07 for flesh (range 0.004--0.22). hydrolysis which converts o, p' and p,p'-DDT to their To our display, we discovered that liver tissue was respective DDE analogs. Base hydrolysis destroys not uniform in regard to DDE, storage. Both lateral TDE. Arochlors and samples containing 0.001 ppm and median lobes from each of four fish were analyzed or more of any of the DDT analogs could be detected and the lateral lobes on the average contained about by this procedure. seven times more DDE than the long median lobes. The presence of other chlorinated compounds, par- Because of this large difference, we had to consider ticularly the polychlorinated biphenyls (PCB's) was separately each type of liver sample. also investigated by microcoulometry after base hy- The values for DDE and DDT in whole albacore drolysis of DDT. The PCB compounds are stable livers, and median lobes are plotted in Figure 6 to under the conditions used. PCB compounds were not show any trends that occurred with time during the found in our albacore samples above our detection summer of 1969. Concentrations of DDE and DDT level of 0.03 2 ppm. werehighly variable (note the log scale). No pro- no'.ncedchanges in the average pesticide level are evi- RESULTS dent during the summer. The slope of regression line. Gamma-ray spectra revealed Zn-65 asthe prominent fitted by least squares to the untransformed pesticide artificial radioisotope present in albacore livers. Ce- values are not significant (P > 0.5) for either DDE sium-137,manganese-54 andcobalt-60werealso or DDT in the median lobes (n = 19), whole livers present. Figure 5 shows that the Zn-65 radioactivity (n = 7), or in flesh (n = 11, not shown in Fig. 6). per gram and the specific activity of Zn-65 of albacore Thus Zn-65 was significantly enhanced while alba- livers increased greatly during the summer of 1969, core resided in coastal waters off Oregon during the thusreaffirming trends observed in previous years by Pearcy and Osterberg (1968). Slopes of linear regres- summer of 1969, but DDT residues did not undergo it sion lines fittedby least squares to both the untrans- similar increase. The lack of positive correlations be- formed and logarithms of radioactivityvalues are tween Zu-65 and DDT or DDE further indicated the highly significant (P <0.01, n = 28). independence ofthesepollutants.Correlationco- Our analyses of DDT residues in albacoreliver and efficients were not significant (P > 0.05) among Zn.65 flesh tissue are summarized in Table 1. The levels of and p, p'-DDE, Zn-65 and p, p'-DDT, and Zn-65 r Since this paper was written, some extractsof whole livers have specific activity and DDT. The correlation between been reanalyzed forPCB's by an improvedmethod. Values ranged from 0.02 to 0.03 ppm using 1260 formulation as a specific activity and DDE, however, was significant standard,Since many of the samples were collected inplastic (P = 0.05). These correlations were based on twelve bags and possibly contaminatedwithPCB's, these values must be suspect. comparisons of livers from individual albacore.

TABLE i DDT residues, parts per million, on a wet weight basis in albacore collected of Oregon, 1969. (8 = equivalent to blank value) PPM

p, p'-DDE o,p'-DDT p, p'-TDE p. p'-DDT

a z Range n x Range n Range a z Range

Wholelivere______7 0.025 0.01- 4 0.005 8-0.01 4 0.007 0.004- 6 0.012 0.001- 0.05 0.01 0.03 Median lobe______19 0.023 0.001- 9 0.009 8-0.020 9 0.004 0.004- 14 0.011 B-0.04 0.14 0.016 Laterallobes---______4 0.00 0.022 __- --- 0 .16 Flesh______11 0.035 0.003- 7 0.012 B-0.028 ______11 0.02 0.001- 0.14 0.04 Blank values (per 3 gn, ssn Ales) ----- 4 0.0002 0.0003- 4 0.0006 0.0002- 4 <0.0004 4 0.0008 0.0W7- 0.0028 0.001 0.003 Percentrecoveries. ______4 93.7 90-113 3 90.3 94-98 4 88.3 83-94 4 90.3 80-93 86

REPORTSVOLUME NVI, 1 JULY 19710 TO30 JUNE 1971 71

DISCUSSION The average values for total DDT residues in albacore were very similar to those reported by Risebrough Our results indicate small amounts of DDT residues (1969) for skipjack tuna(Euthunnus pcl¢nti's)off in albacore tuna and no significant increase in levels Hawaii and the Galapagos Islands. Thus we have no during their sojourn in Oregon waters in 1969, even evidence that freshwater drainage into coastal waters though Zn-65 levels increased drastically as a result is a major source for DDT contamination in albacore. of their association with the Columbia River plume. Our results provide additional support for aerial transport as an important mechanism for transfer of Seasonal variations in the DDT residues of other fishes chlorinated hydrocarbons into the oceans. have been reported (Kelso et al., 1970; Smith and The lack of correlation between freshwater drain- Cole, 1971), indicating that rapid uptake of DDT is age from agricultural lands and DDTin marine fishes possible. Our hypothesis that the Columbia River is a has been reported by Risebrough etat.(1967) and majorsource of DDT for the ocean off Oregon is there- Risebrough (1969). They found that northern an- fore rejected. chovy in San Francisco Bay contained less DDT DDE Median Lobe A DDE Whole Livers o DDT Median Lobe A DDT Whole Livers

0.1

A A 0

A0 0 0 A A 0.01 A 0 0 n

0 0 0 - - - A ------o ------o - --'a 00 - - - - -00 -

0.001 JUNE JULY 1 AUGUST SEPTEMBER FIGURE 6.DDT residues and samples of albacore livers, 1969. The dashed line at 0.002 ppm represents twice our blank values for Is. p=DOT; any values recorded less than this are located on this line. 87

72 CALIFORNIA COOPERATIVE OCEANICFISHERIESINVESTIGATIONS residues that those off southern California. Cox (this and Folsom and Young (1965) found higher concen- volume) found little evidence for elevated DDT resi- trations of other fallout radionuclides offshore than dues in waters around the Columbia River plume. inshore. Thus fallout of some pollutants, including Aerial fallout and global atmospheric transport of DDT, into oceanic regions far from the area of init ial DDT is further evidenced by Dl)T residues in air- dispersal may actually be higher than into coastal seas. borne dust transported by the Atlantic trade winds Some conclusions of the recent Report of the Study (Risebroughet al..1968), by DDT residues in rain- of Critical Environmental Problems (SCEP, 1971, pp. water, even from the agriculturally remote Shetland 133, 135) provide an ideal ending for this paper but Islands (Tarrant and Tatton, 1968), by the occur- hopefully not an epitaph for the oceans. They con- rence of pesticides in Antarctic birds and mammals cluded that " . . DDT and its residues are most (Tatton and Ruziek, 1967), in seals and porpoises in probably distributed throughout the marine biosphere Scotland and Canada far from sites of application ." and "the atmosphere appears to be the major (Ilolden and Marsden,1967),in the Bermuda petrel route for transfer of DDT residues into the oceans, (Wurster and Wingate, 1968), other shearwaters and . and ultimate accumulation site of DDT and its other pelagic birds (Risebroughet al.,1967), and in residues". sperm whales (Wolman and Wilson, 1970). The ocean distribution of some radionuclides that were produced by nuclear explosions and transported ACKNOWLEDGMENTS by world-wide atmospheric circulation also indicate This research was supported in part by the Atomic less accumulation of fallout in coastal waters. Pearcy Energy Commission (Contract AT (45-1) 2227, Task and Osterberg (1968) found that manganese-54 levels Agreement No. 12, RLO-2227-T12-1) and the National decreased with time in albacore caught during the Science Foundation Institutional Sea Grant (G1145). summer off Oregon. This trend suggested higher up- We are grateful to Miss Erika Saunders who assisted take of this fallout radionuclide in oceanic than in in the chromatographic analyses, to William Renfro nearshore waters, a conclusion that was supported by who made helpful editorial comments, and to the cap- the higher amounts of bin-54 found in oceanic than tains and crews of the TYPHOON, SUNRISE, CAYUSE AND coastal zooplankton. Pillai, Smith and Folsom (1964) YAQUINA who provided albacore for analyses. 88

REPORTS VOLUME XVI, 1 JULY 1970 TO 30 JUNE 1971 73

REFERENCES Park, K.1966.Columbia Itiver plume identification by xpe- cific alkalinity. Limnot.Oceanogr. 11:118-120. Barnes, C. A. and M. G. Gross.1966.Distribution at sea of Pearcy, W. G. and C. L. Osterberg.1967.Depth, diet, sea- Columbia River water and its load of radionuclides. In Dis- sonal and geographic variations in zinc-65 of midwater ani- posal of radioactive wastes into seas, oceans and surface mals of Oregon.Int.J. Ocenvol-. Limnol. 1: 10:3-110. waters. Int. Atomic Energy Agency, Vienna 1966, p. 291-3(t2. Pearcy, W. G. and C. L. Osterberg.1968.Zinc-65 and man- Carey, A. G., Jr., AV. G. Pearcy and C. L. Osterberg.1966. ganese-54 in albacore Thunnu..s alalunya from the west coast Artificial radionuclides in marine organisms in the Northeast coast of North America.Linenol,Oceanogr.13:490-498. Pacific Ocean off Oregon. In Disposal of radioactive wastes Pearcy, W. G. and J. L. Mueller.1970.Upwelling, Columbia into seas. oceans and surface waters, Int. Atomic Energy River plume and albacore tuna. Proc. Sixth Int. Symp. on Agency, Vienna. 1966, p. 303-319. Remote Sensing of Environment, pp. 1101-1113. Folsom, T. R. and 1). It. Young. 1965). Silver-110 and cobalt- Pearcy, W. G.1971.Remote sensing and the pelagic fish- 60 in oceanic and coastal organisms.Nature206: 803-806. cries environmentoff Oregcin. Sytnp. on Remote Sensing in Frederick,L.C. 1967.Dispersion of theColumbia River Marine Biology and Fishery Resources,CollegeStation, plume based on radioactivity measurements. Ph.D. thesis, Texas, pp. 158-171. Oregon StateUniv. Library, 109 pp. Pearcy,W. G. in press.Albacore oceanography off Oregon, Gross,M. G., C. A. Barnes and G. K. Riel.1965.Radioac- 1970. Fish Bull. U.S. tivity of the Columbia River effluent.Science149:1088-1090. Perkins, R. W., J. L. Nelson and W. L. IIaushild.1966.Be- Holden, A. V. and K. Marsden.1967. havior and transport of radionuclides in the Columbia River Organochlorine pesti- betweenHanfordand Vancouver, Washington.Limnot. cides inseals and porpoises.Nature 216:1274-1276. O cca n ogr. 11: 23,71-248. Kelso, J. R. Al., H. It. 1lacCrimmon and D. J. Ecobichon. Pillai; K. C., it. C. Smith and T. R. Folsom. 190-1. Pluto- 1070. Seasonal insecticide residue chargesin tissues of fish nium in the marineenvironment. Nature 203:568-571. from Grand River, Ontario. Trans. Amer. Fish. Soc. 99: Risebrough, It. W., I). B. Menzel, D. J. Martin, Jr., and 11. S. 423-426. Olcott.1067. DDT residues in Pacific sea birds: a persist- Leopold, L. B.1962.Rivers. Amer. Sci. 50: 511-537. ent insecticide in marine foodchains.Nature 216:38()-591. Manigold, I). B. and .T. A. Schulze.1969.Pesticides in se- Risebrough, It. W., It. J. Ifuggett, J. J. (irillin and E. 1). Gold- lected western stre:uns-a progress report. Pesticide Muni- berg.1968.Pesticides:'I'rnnsntlnnticmovementsinthe toring J. 3: 124-135. Northeast Trades. Science 1.59: 1233-1236. Osterberg, C., W. G. PearcyandH. C. Curl, Jr.1064.Radio- Risebrough, It. W.1969.Chlorinated hydrocarbons in marine activity and its relationship to oceanic food chains. J. Mar. ecosystems. In Chemical Fallout. Ed. II. W. 'Miller and G. Res. 2.3: 2-12. G. Berg. (',has. C. Thomas I'ultl., U.S.A. pp. 5-23. Osterberg, C., J. Pattnllo and W. Pearcy.1904.Zinc-65 in SCEP, Report of the Study of Critical Environmental Prob- euphausiids as related toColumbiahirer water off the Ore- lems.1971. Man'sImpact onthe GlobalEnvironment. gon coast. Limnot. Oceanogr. 9: 249-257. MIT Press. 319 pp. Smith, It. M. and C. F. Cole.1971.Chlorinated hydrocarbon Osterberg,C.L.1965.RadioactivityfromtheColumbia River. Ocean Science and Ocean Engineering (ASLO-NITS) insecticideresiduesinwinter flounder,Pxeudoplcuronectis 2:968-976. americanus, from theWeiveantlcRiver estuary,Massachu- setts, J. Fish. Res. 13d. Canada 27: 2:374-2380. Osterberg, C. I,., N. CutsballandJ. Cronin.1965.Chromium- Tarrant, K. R. and J. O. Tatton.1968.Organochlorine pes- 51 asa radioactive tracer of Colombia River plume water at ticides inrainwater in the British Isles.Nature219:725-727. sea. Science 150: 1555-1587. Tatton, J. O. and J.II. A. Ituzicka.1967.Organochlarine Owen, R. W.,Jr.1964.Oceanographic conditionsinthe pesticides in Antarctica. Nature 215: 346-348. northeastPacific Ocean and their relation to the albacore Wolman, A. A. and A. J. Wilson, Jr.1970.Occurence of fishery. U.S. Fish Wildl.Serv. Fishhull. 66:503-526. pesticidesinwhales. Pesticide Monitoring J. 4: S-10. Panshin, D.A. 1971 Albacore tuna catches in the northeast Pa- Worster, C. F., Jr. and 1). R. Wingate.1968. DDT residues cific during summer 1.969 as related to selected ocean condi- and declining reproduction in theBermudaPetrel. Science tions. Ph.D. thesis, Oregon StateUniv. Library, 110 pp. 159: 979-981. 89

RLO-2227-T12-2

Reprinted from PACIFIC ScII NCE, Vol. XXVI, No. 3, July, 1972

New Records for Four Deep-Sea Shrimps from the Northeastern Pacific'

ROBERT A. WASMER2

IN 1961 the Department of Oceanography ofW. The specimens are a male with a carapace Oregon State University began a long-term length (postorbital) of 40 mm and a female ecologicalstudy of the northeasternPacific with a carapace length of 53 mm. Ocean off Oregon extending to a distance of Hemipenaeus spinidorsalisis closely related 833 km, generally between lat 43° and 46° N. toH. carpenteriWood-Mason. These species Among the many deep-sea shrimps collected share a number of characters which, according during the course of this study are four species to Ramadan (1938), may entitle them to sub- which were taken in hauls made at positions well generic recognition: they possess a spine at the outside their previously recorded ranges. It isend of the carina of the third abdominal seg- the purpose of this paper to discussthe dis-ment; the three anterior pereiopods have a much tributions of these speciesas they are nowlonger carpus than the other species of the known. All of the specimens have been depos-genus, and their meri are not flattened; they ited in the collections of the Department ofpossess a podobranch on the 12th somite (third Oceanography, Oregon State University., Thepereiopod), an epipodite on the 13th somite collection of the specimens at sea by Depart- (fourth pereiopod), and exopods on all pereio- ment of Oceanography vessels was supported bypods. U.S.Atomic EnergyCommissioncontracts The two specimenspossesscharactersas AT(45-1)1.726 and AT(45-1) 2227, Taskgiven by Faxon (1895) and Burkenroad (1936) Agreement no. 12, RLO-2227-T12-2. which distinguishH. spinidorsalisfromH. car- penteri:rostrum reaching beyond end of eyes ORDER DECAPODA and more than one-fifth the length of the carapace; antennular stylocerite reaching to the SUBORDER NATANTIA tip of externodistal tooth of proximal segment SECTION PENAEIDAA of antennular peduncle; and, in the male, hav- FAMILY PENAEIDAE ing median blade of bipartite appendix mascu- line in the form of a long triangular tooth and Hemipenaeus spinidorialisBate, 1881 shorter than the lateral blade, which is concave within and furnished with setae on its distal Hemipenaeus spinidorsaliswas originally described from specimens trawled by H.M.S.border (see Faxon, 1895, Plate L, Fig. le and 2). Challengerfrom near Tristan da Cunha in the South Atlantic and from near the Philippines inthePacific(Bate,1881,1888). Faxon Plesiopenaeus armatus (Bate), 1881 (1895) recorded the speciesin the eastern Pacific from offCentral America and the Plesiopenaeus armatus was originallyde- Galapagos Islands. It has been taken at depthsscribed from specimens trawled by H.M.S. between 1,867 and 3,749 m (Ramadan, 1938). Challenger from the South Atlantic, the Aus- The two specimens recorded here were taken tralianarchipelago,the North Pacific(lat in a beam trawl to a depth of 3,687 m on 434°37' N, long140°32'E and lat 36°10' N, June 1970between]at44°40.7' N, longlong 178°0' E), and from the South Pacific 133°28.1' W and lat 44°40.9'N, long 133°24.5' (Bate, 1881, 1888). Subsequently it has been recorded,under one name or another (Burken- Manuscript received S October 1971. road,1936,andRamadan,1938, discuss the 2 Oregon State University, Department of Zoology, synonomy of the genus andspecies),from the Corvallis, Oregon 97331. Indian Ocean(Alcock,1901; Ramadan, 1938) 259 90

260 PACIFIC SCIENCIh, Volume 26, July 1972

andAtlantic Ocean(Smith,1884;Faxon, the Gulf of Panama (Faxon, 1893, 1895), and 1896; Bouvier, 1905, 1908; Milne-Edwards andhas subsequently been recorded fromthe Indian Bouvier,1909;Sund,1920;Robertsand Ocean (Alcock and Anderson, 1896; Ander- Pequegnat, 1970). It has been taken at depthsson, 1896; Alcock, 1899, 1901; Balss, 1925) between 750 and 5,400 m. and Atlantic Ocean (Balss,1925; Holthuis, The present specimen was taken in a beam1951; Springer and Bullis,1956; Figueria, trawl to a depth of 3,724 m on 3 June 1970 1957; Fisher and Goldie, 1961; Crosnier and between lat 44°40.2' N, long 133°35.7' WForest, 1968; Foxton, 1970).It has been re- and lat 44°39.5' N, long 133°38.3' W, and so corded from hauls made to 3,200 m. is the first record of the species from the north- Although Forss(1966)figured and dis- eastern Pacific. It is a female with a carapace cusseda damaged specimen ofSystellaspis length of 77 mm (combined rostrum and cara- cristatafrom off Oregon (taken at night from pace length of 127 mm). The specimen agreesa depth range of 1,000=500 in, on 20 October with descriptions of thespecies,possessing 196.4, between ]at 44°28' N, long 125°20' characters given by Burkenroad (1936) andW and lat44°34.5' N,long 125°31.4' W), it Ramadan (1938) which distinguishit fromis nevertheless satisfying to record the species other species in the genus: last four abdominal from the area again. Two additional specimens terga carinate dorsally, a mobile spine on menis have been identified from OSU oceanography of first and second pereiopods, and a strongcollections. The first, a female with a carapace ischial tooth on first pereiopods. length of 11 mm, was taken in an Isaacs-Kidd midwatertrawltowed at night through the SECTION CARIDEA depth range 200-0 m, on 20 July 1961, be- FAMILY OPLOPHORIDAE tween lat46'14.41 N, long 124°40.4' W and lat 46°14.4' N, long 124°33.6' W. A second Acanthephyra micropthalma Smith, 1885 female specimen (carapace length 11.5 mm), Acanthephyra micropthalmawas originallyalso from anIsaacs-Kidd midwatertrawl, was described from specimens trawled from thetaken during a daylight tow on 23 September western Atlantic off the east coast of the United1969, at a station located 65 miles off the cen- States (Smith, 1885, 1886). Elsewhere in thetralOregoncoast(lat44°39.2' N,long Atlantic it has been recorded from off Portugal125°39.8' W), from a depth range of 500-0 by Coutiere (1911)and from southwest of the m. Both specimens agree with the description Azores by Sivertsen and Holthuis (1956). Al-of the species given by Holthuis (1951). cock (1901) recorded it from the Bay of Ben- gal. In the Pacific it has only been recorded from the Celebes Sea and the southern Pacific DISCUSSION by Bate (1888) asA. longidensBate. It has The actual depth of capture of specimens been taken at depths between 2,000 and 4,700taken by open trawls and dredges isalways in m (Sivertsen and Holthuis, 1956). doubt since these may take specimens during The present specimen was taken in a beamdescent and ascent, in addition to the normal trawl to a depth of 3,655 m on 1 June 1970samples taken on the bottom. Because of this, between lat 44°27' N, long 132°14' W andit is proper to ask whether the shrimpsHemi- ]at 44°24.6' N, long 132°12.9'W, and so is penaeus spinidorsalis,Plesiopenaeus armatus, the first record of the species from the north- andAcanthephyra micropthalma,which were eastern Pacific. The specimen, which is a fe-collected only by bottom trawls, are benthic or male with a carapace length of 22 mm, agreespelagic. with the descriptions and figures of the species Examination of forcgut contents from the in the literature. specimens ofHemipenaeus spinidorsalisand Plesiopenaeus armatusshowed the following to Systellaspis cristata (Faxon), 1893 be present: calcareous shell fragments; Radio- Systellaspis cristata was originallydescribed laria; Foraminifera; and unidentified crustacean from specimenstaken by theAlbatrossfromremains. In addition, the foregut contents from 91

New Records for Shrimps from the Northeastern Pacific-WAsMER 261

Hemipenaeus spinidorsalisincludedsmall sec- be exclusively pelagic in habitat (Kemp, 1939). tions of brown tube composed of cemented Examination of the published capture records debris, whereas those fromPlesiopenaeus ar- for this species reveals, however, that all ex- matusincluded fairly large pieces of skeletal cept one are from trawl or dredge samples. The plates and spines from ophiuroidarms, several exception is that reported by Coutiere (191.1) segments of a hollow calcareous tube, and a who recorded the species (as Acanthephyra Ion- large amount of what appeared to be sediment. gidensBate) from a " Bouree net" towed at a The evidence presented by the foregutcon- high rate of speed. The samples collected with tents from these specimens thus indicates thatthis net consisted entirely of shrimps consid- they had resorted to the bottom to feed. Thisered to be pelagic, the vast majority belonging conclusion, along with the fact that all thepre- to the Oplophoridae. The foregut contents of vious capture records for the two speciesare the present specimen consist solely of unidenti- only from trawl or dredge samples, suggestsa fied crustacean remains, with a complete lack benthonicexistenceforbothHemipenaeus of any remains indicating bottom foraging, as spinidorsalisandPlesiopenaeus armatrrs. was the case in the other two species. In addi- Although most adult Penaeidae,whethertion, the pereiopods do not appear to show any neritic or oceanic, are benthonic, a pelagic ex-special adaptations for a benthic existence, but istence does occur sporadically throughout the appear to be similiar to those of other members family, and it is possible that behavior of both of the genus. types is found together in certain abyssal spe- Thus, it can be concluded thatA. micro p- cies of the family (Burkenroad, 1936). Con- thalmais a species which is probably not conceivably then, specimens ofHemipenaeus spini- fined to the immediate neighborhood of the dorsalisandPlesiopenaeus armatusmight swimbottom, but which does show structural evi- up from the bottom at times, so that they mightdences of inhabiting very great depths (very be taken by pelagic hauls. That these speciespoorly developed eyes and soft integument). do have a strong swimming capability issug- The presence of this species in the sample thus gested by their long and well-developed pleo-would be due to its having been caught either pods. as the trawl descended or ascended. If this is Burkenroad (1937).pointed out, however, the case, the vertical distribution of the species that "the termbenthonicwhen applied to Dc- is almost certainly below 1,000 m, inasmuch as capoda Natantia, can refer at most to a vital it has never been taken in the many midwater dependence upon the bottom, rather than to an trawls from 0---1,000 in which have been taken entirely suhstratal existence." If this is accepted, off Oregon. then the apparent obligatory resort to the bottom to feed by the specimens ofHemipenaeus spinidorialisandPlesiopenaeus armatus would LITERATURE CITED bring these species within the range of the term, even though at times they might conceiv-ALCOCK, A. 1899. A summary of the deep- ably be found in the water column. sea zoological work of the Royal Indian Ma- The environmental characteristics of the sed- rine Survey Ship "Investigator" from 1884 iments and bottom water at the station where to 1897. Sci. Mem. Med. Offrs. Army India Hemipenaeus spinidorsalisandPlesiopenaeus 11:1-49. armatuswere taken are as follows (A. G. .1901. A descriptive catalogue of the Carey,personalcommunication) :sediments Indiandeep-sea Crustacea Decapoda Macrura (0.3 percent sand, 62.1 percent silt, and 37.7 and Anomala in theIndian Museum. Indian percent clay); bottom water (1.55°C, 2.98 Museum, Calcutta. 286 p. ml/liter 02i and 34.66 % salinity). ALCOCK, A., and R. S. ANDERSON. 1896. Il- The situation is somewhat different in the lustrationsof the zoology of the Royal In- case ofAcanthephyra mrticropthalma.This spe- dian Marine Surveying Steamer"Investi- cies, together with.SVrtellarpis cri.rtaid,belongs gator," Crustacea, pt. 4, pl. 16 -17. to a fancily (Oplophoridac) which appears to ANDERSON, It.S.1896. An account of the 92

262 PACII-IC SCIENCE, Volunic26, July1972

deep sea Crustacea collected during the sea- .1895. Reports on an exploration off son 1894-95. Natural history notes from the west coasts of Mexico, Central and South H.M. Royal Indian Marine Survey Ship "In- America, and off the Galapagos Islands, etc. vestigator,"CommanderC.F.Oldham, XV. The stalk-eyed Crustacea. Mem. Mus. R.N., commanding, ser. 2, no. 21. J. Asiat. Comp. Zool. Harv. 18:1-292. Soc. Bengal 65 (2) :88-106. . 1896. Reports onthe results of dredg- BALSS, H. 1925. Macrura der deutschen Tief- ing ... in the Gulf of Mexico and the Carib- see-Expedition. 2. Natantia, Teil A. Wissen- bean Sea, and on the east coast of the U.S., schaftliche Ergebnisse der deutschen tiefsee 1877 to 1880, by the U.S. Coast Survey Expedition auf dem Dampfer "Valdivia" Steamer "Blake." Bull. Mus. Comp. Zool. 1898-1899, 20 (5) :217-315. Harv. 30 (3):154-166. BATE, C. S.1881. On the Penaeidea. Ann. FIGUEIRA, A. J. G. 1957. Madeiran decapod Mag. Nat. Hist., ser. 5, 8 (45) :169-196. crustaceans in the collection of the Museu . 1888. Crustacea Macrura. In Report on Municipal do Funchal. I. On some interesting the scientific results of the voyage of H.M.S. deep-sea prawns of the families Pasiphaeidae, Challenger during the years 1873-76. Vol. Oplophoridae and Pandalidae.Bol. Mus. 24 (part LII) :1-942. Her Majesty's Sta- Funchal 10 (25) :22-51. tionery Office, London. FISHER, L. R., and E. H. GOLDIE. 1961. New BOUVIER,E. L. 1905.Sur lea Peneides et les records fortwo deep-sea decapods. Crusta- Stenopides recueillis par les expeditions fran- ceana 2 (1):78-79. caises et monegasques dans 1'Atlantique ori- FORSS, C. A. 1966. The oplophorid and pasiph- ental. C. R. Acad. Sci., Paris 140:980-983. aeid shrimp from off the Oregon coast. Ph.D. -. 1908. Crustaces decapodes (Peneides) thesis,Oregon StateUniv. Corvallis. provenant des campagnes de1'Hirondelleet FOXTON, P. 1970. The vertical distribution of de laPrincesse-Alice(1886-1907). Result. pelagic decapods (Crustacea: Natantia) col- Camp. Sci.Monaco33:1-122. lected on the SOND cruise, 1965. I. The BURKENROAD, M. D. 1936. The Aristaeinae, Caridea. J. Mar. Biol. Ass. U.K. 50 (4) : Solenocerinae and pelagic Penaeinae of the 939-960. Bingham OceanographicCollection.Bull.HOLTHUIS, L. B. 1951. The Caridean Crustacea Bingham Oceanogr. Coll. 5(2):1-151. of tropical west Africa. Atlantide Rep.2:7- ---. 1937. Some remarks on the structure, 187. habits, and distribution of the benthonic KEMP, S. W. 1939. On Acanthephyra purpurea sergestidSicyonelleBorradaile(Crustacea, and its allies (Crustacca Decapoda; Hoplo- Decapoda). Ann. Mag. Nat. Hist., ser. 5_19 phoridae). Ann. Mag. Nat. Hist., ser.11, (49):505-514. 24:568-579. FOREST. 1968. Note CROSNIER, A., and J. MILNE-EDWARDS, A., and E. L. BOUVIER. 1909. preliminaire surlesCarides recucillis par Reports on the results of dredging in the I' "Ombango" an large du plateau continental Gulf of Mexico ... by the U.S. Coast Survey du Gabon a 1'Angola (Crustacea Decapoda XLIV.Les Peneides et Ste- Natantia). Bull. Mus. Hist. nat., Paris, ser. Steamer "Blake." 2, 39 (6) :1123-1147. nopides. Mem. Mus. Comp. Zool. Harv. 28 COUTIERE, H. 1911. Sur les Crevettes Eucy- (3) :181-274. photes recueillies en 1910 au moyen du Filet RAMADAN, M. M. 1938. Crustacea: Penaeidae. Bouree, par laPrincesse-Alice. C.R. Acad. Sci. Rep. Murray Exped. 5 (3) :35-76. Sci., Paris 152:156-158. ROBERTS,T. W., and W. E. PEQUEGNAT. 1970. FAXON, W. 1893. Reports on the dredging Deep-water decapod shrimps of the family operationsoff the west coast of Central Penaeidae, p. 21-57.In W. E. Pequegnat America to the Galapagos .... VI. Prelimi- and F. A. Chace, Jr. [ed.] Contributions on nary descriptions of new species of Crustacea. the biology of the Gulf of Mexico. Gulf Bull. Mus. Comp. Zool. Harv. 24 (7) :149- Publ. Co., Houston, Texas. 220. SIVERTSEN, E., andL. B. HOLTHUIS. 1956. 93

New Records for Shrimps from the Northeastern Pacific-WASMER 263 Crustacea Decapoda (the Penaeidae and Ste- 1886. Report on the decapod Crustacea nopodidae excepted). Rep. SarsN. Atl. Deep of the "Albatross" dredgings off the Sea Exped. 5 (12):1-54. east coast of the United Statesduring... 1884. Rep, SMITH, S. I. 1884. Reporton the decapod Crus- U.S. Comm. Fish. 13:605-706. tacea of the "Albatross" dredgings off the east SPRINGER,S., andH.R.BULLIS, JR. 1956. Col- coast of the United States in 1883. Rep.U.S. lections by the "Oregon" in the Gulf of Comm. Fish. 10:345-426. Mexico. Spec. Sci. Rep. U.S. Fish Wildl. 1885. On some newor little known Serv. 196:1-134. decapod Crustacea, from recent Fish Corn- SUND, O. 1920.Peneidesand stenopides from mission dredgings off theeast coast of the the "Michael Sars" North Atlantic Deep-Sea United States. Proc. U.S. Nat. Mus. 7:493- Expedition 1910. Rep. Sars N. Atl. Deep Sea 511. Exped. 3 (part 2, no. 7) :1-36. 94

reprinted from: RADIONUCLIDES IN ECOSYSTEMS Proceedings of the Third National Symposium on Radloecology, D. J. Nelson, Editor CONF-710501 (1973) RLO-2227-T12-4

ZINC-65 IN OREGON-WASHINGTON CONTINENTAL SHELF SEDIMENTS Norman Cuishall. William C. Renfro,' David W. Evans, and Vernon G. Johnson Department of Oceanography Oregon State University, Corvallis

ABSTRACT Approximately 520 Ci of 6"Zn is estimated to he present in sediments off the Columbia River mouth according to river discharge and measurements of 65Zn concentrations in river water. Analysis of topmost I cm of shelf sediments for 65Zn reveals that 45 Ci is present in this layer within an area bounded by a constant activity'arca contour of 0.5 nCi/m2 (_loto m2). Mean regression coefficients for six vertical profiles yield a half-value depth of 4.5 cm in sediment. Another 270 Ci is therefore estimated to be buried below I cm within the same area. The remaining 205 Ci appears to have been transported beyond the 0:5-nCi/m2 boundary. These observations are compatible with known bottom currents in the shelf.

INTRODUCTION PREDICTED OCEAN INVENTORY

Two principal routes of entry of "Zit into the Pacific If a constant rate of delivery of 65Zn into the sea had Ocean are known: worldwide fallout from atmospheric persisted for a few years. it would be a rather simple nuclear detonations and input, via the Columbia River. matter to calculate the resulting steady-state inventory. from reactors at Hanforo. Washington. The latter source Transport rates downstream vary seasonally with river is considered to predominate within a few hundred discharge.however. and have also declined during kilometers of the river mouth. The clear association of recent years owing to reactor shutdowns. During 1968, 6sZn content of marine organisms with the known transport rates pastAstoria. Oregon. near the river position of the Columbia River plume strongly supports mouth, varied from 0.5 Ci/day (December) to 8.2 this view (Osterberg. Pattullo. and Pearcy 1964). In Ci/day (June), with an average of 2.8 Ci/day (see addition, the 65Zn specific activity (65Zn/total Zn) in Appendix for discussion of calculations). The average the Columbia River system clearly marks Columbia during 1969 was 2.3 Ci/day. When monthly discharges River zinc. Whereas 65Zn specificactivities found during 1968-69 are cumulated by month and allow- elsewhere are in the range 10-2 -10' pCi/g (Alexander ance is made for radioactive decay, the resultant total and Rowland 1966. Wolfe 1970), values of 103-106 inventory in the marine environment is 550 Ci on I pCi/g are routinely observed near the Columbia River. January1970. Transport rates prior to1968 were The analyses reported in this paper become practically greater. but their cumulative total is subject to consider- insensitive when the specific activity is below about 3 X ably more decay. With a 245-day half-life 65Zn decays 103 pCi/g. Thus they represent only "Zn from the to A/Ao = 0.127 intwo years. We estimate the Columbia River. Zinc-65 has previously been reported pre-1968 residual contribution to be 100 Ci onI in marine sediments near the river mouth (Osterberg. January 1970 for a total inventory of about 650 Ci. Kulm, and Byrne1963. Gross1966). Indeed, the Not all65 Zn is associated with sediment. In the lower sorption and retention of 65Zn by sediments in the river,filtrationof water through 0.45-p membrane river proper (Johnson 1966) suggest that the principal filterstypically removes 75to8557, of "Zn. The reservoir of Columbia River zinc in the sea would be remainder is presumably in "solution" either as ions, marine sediment.Thepresent work represents an complexes, or perhaps on (in) colloids small enough to attempt to assess the size and geographic shape of this pass through the filter. Lowman et al. (1966) found reservoir. Observed distribution and quantity of 65Zn that essentially all "dissolved" Zn in river water from inshelf sediments willbe compared with atotal the west coast of Puerto Rico rapidly becomes associ- inventory predicted from river discharge data. Alterna- ated with particles after entering the sea. If similar tive extreme models based on burial and on transport of events occur off the Columbia River it is possible that 65Zn will be considered. Finally,we will attempt to virtually the entire 650-0 ocean inventory is contained construct an accountability ledger for 65Zn. in suspended and bosom sediments. On the other hand, 1. Present address: International Laboratory of Marine Radio. if no such precipitation occurs in the Columbia River activity. Musee Occanographique. Principality of Monaco. system. the sediment inventory might he only 80% of

694 95

695

Table 1. Zinc-65 content of Columbia River sediment exposed to seawater

Exposure Total 6! Zn content CuSO4 leached. "s Zn 65 Zn leached with CuSO4 time (days) (pCi/g ± counting error) (pCi/g ± counting error) (percent ± counting error)

0 69.2 ± 6.6 57.6 ± 10.2 35.7 ± 1.6 79.1 ± 6.8 66.2 ± 5.6 68.1 ± 6.0 28.6±6.0 41.9±9.6 72.2±5.3 (D) 29.4±2.2(D) 40.7±4.3 12 69.4 ± 5.7 33.9 ± 2.0 48.814.9 69.26.0 (D) 32.5 ± 1.2 (D) 46.8 ±4.2 26 66.8 ± 8.8 35.2 ± 5.3 52.6 ± 10.5 71.3 10.0 (D) 35.2 ± 5.8 (D) 49.3 ± 10.7 71.5 11.3 (D) 28.2±5.3(D) 39.4±9.7 50 65.3 ± 7.1 24.7 ±4.1 37.8 ± 7.5 62.1 7.1 (D) 32.7 ± 1.6 (D) 52.6 ±6.5 65.07.2 (D) 39.6 ± 1.9 (D) 60.9±7.4 77 71.7 ±6.6 (D)

Note:Samples marked (D) were contained in dialysisbags.All values have beencorrected for radioactive decay. the total or 520 Ci. We will use the latter value and Periodic samples of each portion were collected at considerita minimum. We also assumethatthe intervals over the following II weeks. Each of these inventory remained constant throughout 1970. samples wasanalyzed by direct counting and by the If substantialdesorption or release of particle-bound CuSO4 leaching method described later. Bay-water 6sZn occurred in seawater the portion of the total salinity at the site was determined twice during the inventorycontained in sediment might be lowered. experiment. On day 12 the salinity was 33.77c and on Johnson, Cutshall. and Osterberg (l%7) reported that day 26, 31.5% These values confirmed the expectation seawater leached only a very slight proportion (about of essentially undiluted seawater at the site. 117G) of "Zn from Columbia River sediments. Their Zinc-65 content of the sediment (Table 1) did not experiments provided only about 2 hr of contact with perceptibly change. except for decay, during I I weeks seawater,however. Furthermore, other authors have of exposuretoseawater. Uncertainties estimated from noted substantialdesorption of metals that have a counting statistics are about 10% (coefficient of varia- chemistry somewhat similartozinc. For example, tion), so that desorption of a few percent might not Murata(1939) found that appreciable manganese is have been detected. It is clear, though, that any losses releasedfrom river-borne sediments to seawater, and from deposited sediments aresmall enough to be Kharkar, Turekian, and Bertine (1967) reported that neglected in estimating the sediment inventory of6SZn. sediment-sorbed cobaltissubstantially desorbed by Thereforeat least520 Ci of 65Zn ought to have been seawater. We have therefore measured the 65 Zn content present in sediments off the mouth of the Columbia of ColumbiaRiver sediment exposed to seawater for River on I January 1970. relatively long times. To approximate losses of 6 s Zn from sediment depos- COPPER SULFATE LEACHING METHOD ited at sea, severalkilograms of sediment (silty sand) collectedfrom the Columbia River near McNary Dam Johnson et al. (1967) noted that solutions of 0.05 M were placed in Yaquina Bay at Newport, Oregon. CuSO4 leached as much as 54% of65 Zn from Columbia Yaquina Bayis about 180 kni south of the Columbia River sediment. We have used CuSO4 extraction for River mouth and its "Zn content is negligible. One analysis of 65 Zn in sediments. Approximately 250 cc of portionof sediment was placed in a shallow plastic pan wet sediment are suspended in 500 nil of 0.05 MCuSO4 and coveredwith a plastic screen which allowed free solution for 1-3 weeks. The suspension is then filtered passage of seawater but excluded large organisms. The and thefiltrate isanalyzed for zinc by atomic absorp- other portion was packaged indialysis bags which tionspectrophotometryandforradionuclides by presumably allowed penetration only by ionic solutes. gamma-ray spectrometry. After filtration the volume of 96

696

filtrate is measured and a subsample is taken for zinc experiment. That is, 65Zn and zinc were extracted at determination. In most cases the zinc concentrations the same rate. arc corrected for CuSO4 blanks, the correction usually amountingtoless than1%. The remainder of the ANALYSIS OF SHELF SEDIMENTS filtrate is evaporated to dryness and the residue placed in a 12-cc tube for counting in a well-type Nal(T1) During 1970, samples were collected on each of three detector.The method allows extractsrepresentifig cruises, in January, March, and December, at locations rather large samples to be placed in a high geometry, shown in Fig. 2. Sediments ranged from sand to silt in thus increasing analytical sensitivity. In addition, com- the area sampled. The topmost') cm was skimmed from monly interfering gamma emitters such as 46Sc, 00K, 250 cm2 of Smith-Mclntyre grab (Smith and McIntyre and 2)4 Bi (138U series) are not extracted. The princi- 1954) samples and immediately placedin CuSO4 pal disadvantage as compared with direct counting is solution. The suspension was returned to the labora- the uncertainty introduced by extraction yield varia- tory at Corvallis. where it was filtered and analyzed as tions. (Specific activity determinations are probably not above. Combined data from all three cruises have been affected by absolute yields, however.) 5. i2.. 2 Sediments from the long-term seawater exposure .26 above were also analyzed by CuSO4 leaching. Extrac- 0 % tion yields (Table I) ranged from 38 to 61%, with a A .o'+r mean of 47% and a standard deviation of ±7.2% for ten extractions. The extraction efficiency does not appear o LA rtRr_`Ltd to change withtimeof exposure of the sediment to A seawater. u We have alsoexamined the rate of extraction of 65Zn to assure that one week contact time is sufficient for maximum yield. Approximately 2 kg of sediment was A 0 0A0 suspended in 16 liters of 0.05 3/ CuSO4 solution. Small A 0 aliquots (175 ml) were extracted at times up to 16 days A WASH later and analyzed for zinc and for gammaemitters. 0 Extraction curves for 65 Zn (Fig. I) rise quite rapidly at F0 o A GRAYS first and become essentially flat after about 3-4 days. /\ HARBOR Final yields in the two runs were 46.4 and 46.6%, in 0 goodagreement with the earlier determinations. Zinc- f'18ArWILL APA 65 specific activity in solution did not varyduring the

4r 1A t 0 O n COL 11M8,A 0oB8 A RIVER Pep ik A o 0 os 0

A JAN.1970 a MAR. 1970 to 0 o DEC. 1970 l o rj ORE

0 El 4 6 8 10 14 12 LNtwroRt TIME /days! 126. '2Y i2.' .2 Fig. 1. Percent of 6Zn removed from sediment by 0.05 M CuSO4 solution as a function of contact time. Fig. 2. Chart showing sampling locations. 97

697 used to construct contour maps for specific activity and activity (Table 3). No yield correction has been made in activity per unit area ( Fig. ; ). Variations in Zn values Fig. 4 and Table 3.If 47% is taken as a uniform at certain locations sampled on different dates have extraction yield, the 21 Ci total in Table 3 represents been noted, so that combination ofall data is not 45 Ci. The topmost Icm of sediment within the rigorouslyvalid.The aim hereisto make afirst 0.5-nCi/m2 contour in Fig. 4 then contains 8.7z of the approximation tothetotal inventory. Variations in calculated 520 total sediment inventory. This is sonie- time will be consideredhen we have series of data for what less than the 14', calculated by Gross tin mess) longer periods. Results of several replicate analyses for1963, although both estimates areonly small (different grabs) indicate surprisingly good analytical fractions of the total. We have considered the extreme precision (Table 2). Areas between contour intervalsinFig. 4 were Table 2. Results of analyses of replicate grab samples determinedgraphically and multiplied by the arithmetic for 65Zn using CuSO4 method mean of contour boundaries to obtain estimates of total 6sZn content

Location . Specific Date Activity/unit N latitude W longitude area activity (nCi/m2) (nCi/g Zn)

46°17' 124°17' 1/70 11.8 76.5 13.0 93.4 12.1 63.3 7.8 52.1

46°14' 124°15' 3/70 4.4 68.2 3.7 63.0 3.5 68.0

46°19' 124°17' 3/70 3.4 40.0 3.1 31.2 4.2 35.7

46°24' 124°19' 3/70 4.1 39.1 4.6 63.8 5.3 66.4

46°31' 124°20' 3/70 5.8 55.9 5.2 66.9 5.4 89.1 46°37' 124°10' 3/70 2.0 70.2 2.3 17.3 2.5 37.4

46°36' 124°16' 3/70 4.6 67.6 4.5 65.0 4.0 62.6

Table 3. Estimation of 65Zn contained in topmost I cm of shelf sediment

Mean 61 Zn/unit Contour Area Total 61Zn area interval (m2) (CO (nCi/m2) (nCi/m2)

>10 1L0 3.7X108 4.0 >4. <10 7.0 5.5 X 108 3.8 >2, <4 3.0 11.9 x 108 3.6 >1, <2 1.5 18.3 x 108 2.7 126- 25. 'a.. it >0.5, <1 0.75 94 X 108 7.0 Fig. 3. Isoplcihs of 65Zn specific activities in topmost I cm Total z 21.1 of shelf sediments, nCi/g in. 98

698

Depth in Sediment, Cm Fig. S. Vertical profiles of 65Zn in sediment. Solid line indicates profile calculated from total-burial model.

area and if no postdepositional vertical mixing occurred. The 45 Ci found within the 0.5-nCi/m2contour is taken torepresent the integral of the activity-depth curve from 0 toI cm deep and the profileis calculated (Fig. 5). The half-value depth for this line is 7.6 cm. Vertical profiles have been determinedat sites designated by letters A through F in Fig. 2. Successively deeper1-cm layers were collected separatelyfrom grab samples andanalyzed. Zinc-65 specific activityfor these samples declineswith depthas expected (Fig.5). However, with the exception of profile C.the slopes are all greater than predicted by the total burial model. Half-value depths based on least-squares fits to profiles A through F are 2.8, 5.0. 35, 3.2, 3.8. and 5.6 cm Fig. 4. Isoplethsof 6SZn in topmost I cm of shelf sediments. respectively. Field notes indicate that sample C was Activityper unit area,nCi/m2 (not corrected for extraction collected from a sampler only partially filledwith sand. yield). It is possiblethat mixing of the sample occurred while the samplerwas being raised. The mean regression casesof two models which would account for the coefficient forthe logarithm of specific activity vs remainder by (1) assuming that the remainder isall depth is 0.154 cm-' (mean of six profiles), which buried below 1 cm but contained within the 0.5-nCihn2 corresponds to a half-depth of 4.5 cm.It is apparent contour, or (2) assuming that the remainder has been from the profiles that some 65Zn is buried below I cm, entirely transported outside the 0.5-nCi/m2 contour. but burial alonedoes notaccount for the totalsediment inventory. BURIAL AS THE EXTREME CASE TRANSPORT AS THE EXTREME CASE Inthis case we assume thatthetotal 520 Ciis contained within the 0.5-nCi/m2 contour and that the Alternatively, we assume that no 65Zn is buried activity-depth profileisexponential. Such a profile below I cm on theshelfbut,rather,isentirely might be expected if both sediment and 65Zn deposi- transported outside the 0.5-nCi/m2 boundary. Zinc-65 tion rates were uniform and constant over the entire is assumed to spread in two dimensions along vectors 99

699

Table 4. Estimates of transport rates for shelf sediments

Rate ikm/year) Method Northward Westward Reference alongshore offshore

65Zn content of sediment (minimum rate) 20-25 12 Gross and Nelson 1966 6SZn:6OCo ratio 30 10 Gross and Nelson 1966 No-burial 6SZn model 3050 425 This work 65 Zn specific activity (minimum rate) 123 20 This work Seabed drifters (bottom current measurement) 365-730 Gross, Morse.and Barnes 1969 radiating from the river mouth. The transport rate is as previouslyreported by Gross (1966). Although 6sZn takento be constant along each vector and proportional isspreadacrossthefullwidth of theshelf off to thedistance from the river mouth to the 0.5-nCi/m2 Washington, no radionuclide evidence yet available contourline.The "age" of sediment (time since would indicate transport onto the slope or the abyssal entering the sea) at this line would then be only 32 plain. days. Transport rates between the river mouth and the Only a relatively small fraction of the total sediment 0.5-nCi/m2 contour would amount to 3050 km/year inventory of 6SZn predicted for the ocean from input (10 cm/sec) northward parallel to the coastline and 425 data is found in the uppermost I cm of shelf sediment. km/year (1.4 cm/sec) westward offshore. This north- Neither the extreme cases of complete burial nor total ward rateis in the range of reported surface currents transport outsidethe study area accounts forthe (Gross and Nelson 1966) and is surely an upper limit. remainder when taken alone. In the complete burial Other means for estimating mean transport rates case, profiles of65 Zn vs depth of burial would need to basedon radionuclides have been used. For example. have lower slopes than are observed. In the no-burial Grossand Nelson 119601 used 6 5 Zn concen tration in case. transport rates would have to be improbably high- sediments and 65Zn:60Co activity ratios in sediments It seems likely that some combination of burial and for such estimates (Table 4). A fourth method assumes transportoutside the 0.5-nCi/m2 contour is needed in that changes in specific activity of 65 Zn are due only to order to account for the total sediment inventory of decay of6SZn and that input specific activities have 61 Zn. Observed "Zn profiles confirm that significant been constant. This method yields a minimum transport burialof 6SZn does occur. This burial cannot represent rate (maximum age). since dilution of Columbia River simple deposition. however. The mean deposition rate sediment with shelf sediment will also lower specific inferred from 65 Zn specific activity profiles in sediment activity by adding stable zinc. Transport rates calcu- is 6.7 cm/year. If extended over the area within the lated using specific activity or the no-burial model are 0.5-nCi/m2 contour, ^1010 m2, this would require the considerably higher than those given by Gross and supply of 6.7 XI0t tkg/year (1.8 X106metric Nelson (1966) (Table 4). tons/day), assuming a water-free, in situ density of 1.0 Bottomcurrent speeds on the Washington shelf have g/cc. The Columbia River discharged an average of 2.3 been estimated by Gross, Morse, and Barnes (1969). X 10° metric tons/day during 1963 (Uaushild et al. They report that bottom drifters move onshore when 1966). Any other source of sediment would cause released where depths are less than 40 m and northward excessivedilution of 6SZn and is therefore ruled out. at greater depths."Typicalspeeds fortheselong- There does not appear to be sufficient sediment supply distancenorthward movements werefront I to 2 to account for6 5Zn burial. We conclude that vertical km/day, ranging from 0.4 to 3.2 km/day." (One to two mixing, either by currents or by organisms, within the km/dayequals 365 to 730 km/year. Four-tenths to 3.2 upper few centimeters of sediment must be responsible km/day equals 150 to 1200 km/year.) for 6SZn atdepths greater than I cm.

DISCUSSION ZINC-65 ACCOUNTABILITY

Columbia River 6SZn and. by inference, Columbia Only 8.7% of the calculated sediment inventory of River sediment are transported predominantly north- "Zn has been accounted for by direct analysis of the ward alongthe continental shelf after entering the sea topmost I cm of shelf sediment off Oregon and 100

700

Washington. Another portionis buried atdepths below especially grateful to Dr. M. C. Gross for the opportunity to read his paper "Sediment-Associated Radio- 1 cnt.If the half-depth calculated from the mean nuclides from the Columbia River, 1961 - 1963" prior regressioncoefficientsofsixprofiles (4.5 ctn)is representative of the entire area inside the 0.5-nCi/m2 to its publication. contour, another 52`%, is buried at depths greater than I This workwas supportedby U.S.Atomic Energy is cm. Thisleaves39% of the total which must have been Contract No. AT(45.1 )2_2__27. Task Agreement 12, and transported outsidetheareabounded by the 0.5. assigned publication numberRLO-2227-T12-4. nCi/m2 contour. When thisnew estimateof the fraction contained LITERATURE CITED contour (610)isused to withinthe0.5-nCi/m2 Alexander. G. V., and R. H. Rowland. 1966. Estimation estimate age at the contour and transport rates, more of zinc-65backgroundlevelsformarine coastal reasonable values are obtained. The calculated age at waters. Nature 210(5032): 155-57. 330 days, and the corresponding the contour is then Gross, M. G. 1966. Distribution of radioactive marine northward alongshoretransport rate is 290 km/year. sedimentderivedfromtheColumbiaRiver.J. very favorably with seabed drifter data which compares Geophy. Res. 71(8): 2017.2 1. (Table 4). Gross, M. G. Sediment-associated radionuclides from the Columbia River.1961-1963. in:Bioenviron- RECAPITULATION AND POSSIBLE mental Studies of the Columbia River Estuary and FUTURE STUDY Adjacent Ocean Region. U.S. Atomic Energy Com- mission. (In press). The regime of sediment transport on the continental Gross, M. G., B. Morse. and C. A. Barnes.1969. shelf has beengreatly oversimplified. Many assumptions Movement of near-bottom waters on the continental have been necessaryin order to allow manipulations of shelf off the Northwestern United States. J. Geophy. availabledata. These assumptions range from probably Res. 74(28): 7044-47. valid, through dubious, to obviously false. The extreme Gross. M. G., and J.L. Nelson.1966. Sediment cases of no deposition and total deposition lead to movement on the continental shelf near Washington improbable conclusions. However, when the extreme and Oregon. Science 154: 879-85. cases are combined to yield an intermediate model, Haushild. W. L., R. W. Perkins. H. H. Stevens, Jr., G. R. inferred6SZn distribution on the shelf is made more Dempster, Jr.. and J. L. Glenn. 1966. Radionuclide and transport rates arein gratifying agree- reasonable transportinthe Pasco to Vancouver. Washington, ment with seabeddrifter data. reach of the Columbia River July 1962 to September Refinementof the absolute calculated numbers is 1963. Portland, Ore. U.S. Geological Survey Openfile possiblewith improved Columbia River 65Zn discharge Report. 187 pp. data and with a broader set of vertical profiles on the Johnson. V. G. 1966. Retention of zinc-65 by Colum- shelf. Allowance for 65 Zn retained in the 14 miles of bia River sediment. M.S. Thesis. Oregon State Univer- estuary between Astoria and the river mouth would also sity. Corvallis. 56 pp. affect thecalculatedvalues. Sorting of different particle Johnson, V. G., N. H. Cutshall. and C.L. Osterberg. sizes during transportmay be indicated by analysis of 1967.Retentionof zinc-65by Columbia River various particlesizes of sediment for 65Zn specific sediment. Water Resources Research 3(1): 99-102. activityalonga north-south transect. Continued meas- Kharkar. D. P., K. K. Turekian, and K. K. Bertine. urement of61Zn in shelf sediments during declining 1968. Stream supply of dissolved silver, molybdenum, input by the rivermay place further constraints upon antimony, selenium. chromium, cobalt, rubidium, and transport considerations. cesium tothe oceans. Geochim. et Cosmochim. Acta. 32: 285-98. ACKNOWLEDGMENTS Lowman, F. G.. D. K. Phelps, R. McClin, V. Roman de We appreciate sampling assistance from the officers Vega. 1. Oliver de Padovani, and R. J. Garcia. 1966. and crew of R/V YAQUINA and many graduate Interactionsoftheenvironmental and biological assistants. Messrs.I.L. Larsen and J. J. Wagner have factors on the distribution of trace elements in the carefully and reliably accomplished the radionuclide marine environment, pp. 249-65 in:Disposal of and stable-zinc analyses. Mr. W. E. Bishop assisted in Radioactive Wastes into Seas. Oceans. and Surface the determination of65 Znextraction rates. We are Waters. IAEA, Vienna. 101

701

Murata, K. J. 1939. Manganese in river and ocean muds. (2) Amer. Jour. of Science 237: 725-35. Osterberg, C. L., L. D. Kulm, and J. V. Bynte. 1963. A Gamma emitters in marine sediments near the Colum- bia River. Science 139(3558): 916-17. and thetotal activity present at steady state (x -. oe) is Osterberg, C. L.,J.Pattullo, and W. Pearcy. 1964. Zinc-65 in cuphausiids as related to Columbia River water off the Oregon coast. Linmol. Oceanogr. 9(2): At=AsS. (3) 249-57. Smith, W., and A. D. McIntyre. 1954. A spring-loaded is bottomsampler. J. Mar. Biol. Assoc. 33(l): 257-b4. The fraction of total steady-state activity which U.S. Geological Survey. Pacific Northwest water re- found above depth x is sources summary. Water Year 1968, Water Year 1969. Ac Wolfe, D. A. 1970. Levels of stable Zn and 65 Zn in Fx=A = (4) Oassotrea virginica from North Carolina. J. Fish. Res. At Bd. Canada 27(I ): 47 -57. Solving Eq. (4) for x,

APPENDIX X=-A. in (I -F.). (5) In order to conservespace and maintain continuity in the above text we have omitted many equations and calculations. These are briefly summarized below: The half-value depth x4 is the depth where Fx = 0.5, or Zinc-65discharge. Monthly discharg e of 65Zn by the ColumbiaRiver was estimated by multiplying single monthly concentrations at Astoria (unpublished data) X/=--5In (I -0.5). (6) by meanmonthly water discharge values (U.S. Geologi- calSurvey). Total 65Zn discharged during each month Combining (5) and (6) and solving for x,A, was corrected for decay toI January 1970 using the decay equationA = A0 e-x". Zinc-65 discharge values I n 0.5 thus estimated may be in error, because both the X (7) concentrationof 6SZn and the water discharge rate Xr/: In (I - Fx) vary ona time scale of days. Greater sampling frequency wouldimprove accuracyof thedischarge The value of F determined for x = I cm is 0.087, so estimates. that Total-burial model. Zinc-65 of constant specific activ- In 0.5 ity is considered to deposit at a constant rate for a 11 sufficientlylong timethat decay of the deposited In (I - 0.087) activity equals the deposition rate. In addition, the sediment deposition rate, S, is also considered to be No-burial model. In the above case we have equated constant, so that depth below the surface is equivalent time with the ratio of depth in sediment to sedimenta- to the product of time since deposition and sediment tion rate. In the no-burial case we treat time as "age" or deposition rate. The activity concentration (or specific elapsed time since entering the sea and further assume activity) at depth, x, is then related to the surface that isopleths of activity per unit area are also isopletls activity concentration by of age. Age or tis then equivalent to A/S in Eq. (4), so that c-at = A. e-(A/S)x e-xt) (8) Ax = Ay (1) F. = (I -

Solving for t, The cumulativetotal activity above depth x is -in (I -- Fx) A,: = f Ase-r,(x/S)dx A 102

702

-ln(1 -0.087) (9) cm-1): = 32 d. 2.83 X 10-3 d-' Ac At = I -C -(a/5)x Transport rates are then calculated by dividingthe distancebetween the river mouth and the 0.5-nCi/m2 450 contour by the age, t. = = 315 Ci. I - e-°' 54cm-' 1 cm Zinc-65accountability. Inthis caseEq. (4) is rearranged so that the total activity within the 0.5-nCi/m2 This is 315/520 or 0.61 of the estimated total sediment boundary (At) can be estimated in terms of theactivity inventory of 65Zn, andthe fraction buriedbelow I cm found in the upper 1 cm (Ae = 45 Ci) and the is 0.61 - 0.087 = 0.523. The new transport rate is experimental mean regressioncoefficient (X/S = 0.154 calculated as in the no-burial model. 103

reprinted from: RADIONUCLIDESIN ECOSYSTEMS ProegedIngs of the ThirdNational Symposium on Rfkdloecology, D. J. Nelson, Editor CONF-710501 (1973)

RLO-2227-T12-5

ZINC-65 SPECIFIC ACTIVITY INMYTILUS CAL/I-ORNIANUS TISSUES' IngvarL. Larsen, William C.Renfro,2 and Norman FI: Cutshall. Department of Oceanography School of Science Oregon State University, Corvallis

ABSTRACT

Zinc-65 specific activity (nCi 65Zn/g Zn) in soft parts of mussels living along the Pacific Northwest coast varies with distance from the Columbia River mouth. At any given site the specific activity in different tissues (muscle, gills, mantle, gonads. foot, and viscera) is less variable than either a measure of radioactivity or stable zinc concentrations. No significant uniform relationship between specific activities in different tissues is evident. Mussels 174 km south of the river mouth have higher 65Zn specific activity than would be predicted by linear mixing of river water with seawater.

INTRODUCTION 8 to 13 cm werecollected from each station. The samples werefrozen and returned to the laboratory to Zinc-65, produced by neutron activation of material be dissected into specific components: adductor mus- present in Columbia River water used to cool the cle, foot, mantle, gills, reproductive organs. and viscera. plutonium production reactors at Hanford,WVashington, The pooled organsweredried,ashed in a mufflefurnace is found in marine organisms widely distributed along at 450 C. and dissolved in 8 of nitric acid. Diluted the Pacific Northwest coast. The California sea mussel, aliquots of this dissolved ash were analyzed for6-1 Zn in Mytilus californianusConrad, has previously been used a 12.7 X 12.7 cm Nal(TI) well-type detector coupled to as a biologicalindicator of 6 5 Zn levels (Watson et al. 1961, Folsom et al. 1963, Nagaya and Folsom 1964, Seymourand Lewis 1964. Young and Folsom 1967, Alexanderand Rowland 1966, Toombs and Culter 1968). Early determinations of this radionuclide were concernedprimarily with its concentration in a given mass oftissue. However, as advocated by the working committee onoceanography (National Research Coun- cil 1962) a more satisfactory index of 65Zn levels in organismsis the specific activity, which expresses in a ratio the amountof a radioactive isotope to the total isotopes ofthat element. Whereas many radioactivity measurementson mussels have been made, few determi- nationsof the 65Zn specific activities of total soft tissue or internalorgans have been reported. This study was undertakento determine 6 5 Zn specific activities in various organsof mussels along the Oregon coast to better assessthe influence of the Columbia River on intertidalorganisms.

COLLECTION AND PREPARATION

Musselswere collected from three stations along the Oregoncoast (Fig. 1) during late June and early July 1967. Twelve to sixteen mussels ranging in length from

1. This research supported by USAIiC contract number AT(45-1)2227, Task Agreement 12, Publication number RLO- 2227-112-5. 2. Present address: International Laboratory of Marine Radio- Fig. 1. Collection areas (Tillamook [lead, Y2quin3 Head. and activity, Mucec Occanugraphique. Principality of Monaco. Cape Arago) and surface salinities off Oregon 5-I5 July 1967.

747 104

748

0.51 0.89 1.12 1.46 ENERGY. Me.. Fig. 2. Gamma-ray spectra. (1) Neutron activated sample 145 days after end of bombardment (normalized to a 400-min count). (2) Spectrum of sample prior to neutron activation (400-min count). (3) 400-min background count.

a 5I2-channel analyzer. Stable zinc was measured by concentration by the stable zinc concentration and neutron activation. The samples were placedin a multiplying by 103 to obtain units of nanocurics of rotating rack in the TRIGA reactor at Oregon State 65Zn/g Zn. The uncertainty term (given at the 95% University and activated for 2 hr with a thermal confidence level) is expressed as a percent of the mean. neutron flux of 6 X 10t t neutrons/cm2/sec. Samples, Thistermincludesuncertainties duetocounting, standards, and reagent blanks were activated simultane- weighing, and comparisons with the activated standards. ously. Following activation the samples were allowed to Samples collected near the mouth of the Columbia "cool" for several months to allow short half-life River have higher 6 5 Zn content and 6 s Zn specific radionuclides to decay. Samples were then counted and activity than samples more distant. Significant uniform the data reduced by a nonlinear least-squares program patterns among tissues for the various stations are not utilizing a 3300 CDC computer. Figure 2 illustrates evident. Values of the 65Zn specific activities are less gamma-ray spectra of a sample counted before and variable than with either 65Zn or stable zinc concentra- several months after neutron activation. tions. Tissues which are high in 65Zn likewise have high stable zinc concentrations. Similarly, tissues low in RESULTS OF ANALYSIS 6tZn also reflect correspondingly low stable zinc. Table 2 presents 65Zn specific activity values deter- Table I summarizes the results of the determination mined by others. Values from Table I corresponding to of 65Zn and stable zinc for the samples. Specific similar areas appear to be in general agreement with activity of 6 $ Zn was calculated by dividing the 6 5 Zn these values.

Table 1. Concentrations of 6SZn and stable Zn in mussels collected from the Oregon Coast

Collection Reproductive Muscle Foot Mantle Gills Viscera area organs

Tillamook Sp. act. nCi 65Zn/g ?_n 785 ± 11.6 835 ± 10.2 759 ± 14.9 798 ±8.45 909 ± 6.22 766 ± 10.4 Head 65Zn pCi 65Zn/g ash 413 ± 8.84 465±6.35 123±12.9 443±3.91 571 ± 1.91 441 ±6.12 (TH) Stable Zn 'pg Zn/g ash 526 ± 7.46 557 ± 7.94 162 ± 7.34 555 ± 7.49 628 ± 5.92 576 ± 8.38 Yaquina Sp. act, nCi 65Zn/g Zn 140±22.8 176±36.4 163±32.1 186±15.5 231 ±13.0 206±16.3 Head 65Zn pCi 65Zn/g ash 72.8 ± 21.8 42.0 ± 35.9 20.4 ± 30.0 37.1 ± 13.4 78.1±9.17 84.8±13.8 (YH) Stable Zn jig Zn/g ash 520±6.65 239 ± 5.83 125 ± 1 1.3 200 ± 7.90 338 ± 9.20 411 ± 8.66

Cape Arago Sp. act, nCi 65Zn/g Zn 50.1 ±88.6 75.0 110 77.0±63.8 50.6 ± 76.9 116±18.5 72.3±21.9 (CA) 65Zn pCi 657.n/g ash 27.3±88.4 18.9±110 12.7 ± 63.5 7.64 ± 76.4 30.6±17.4 69.4±20.4 Stable 7.n pg Zn/g ash 545 ± 5.64 252 ±8.95 165 ±6.38 151 ±9.02 263 ±6.32 961 ±8.10

Note:± indicates the estimated percent uncertainty in the measurementsat the 95% confidence level. 105

749

Table 2. Zinc-65 specific activities in total soft tissues of ,tfviilus califonrianus.

Approximate distance south 65Zn specific Location front Columbia Sampling date activity eference River (nCi 65 Zn/g Zn) km mi

Mouth, Columbia River 0 0 August 1964 706 Alexander and Rowland 1966 Tillamook liead, Oregon 31 19 March 1964 506 Oregon State University (unpublished) August 1964 278 Alexander and Rowland 1966 Cannon Beach, Oregon 40 25 July 1969 308 Toombs 1969 (personal communication) Nehalem River Jetty, Oregon 64 40 July 1969 466 Toombs 1969 (personal communication) Agate Beach, Oregon 172 107 August 1969 147 Toombs 1969 (personal communication) Yaquina Head, Oregon 174 108 March 1964 t35 Oregon State University (unpublished) Cape Arago, Oregon 335 208 March 1964 670 Oregon State University (unpublished) La Jolla, California 1770 1100 1964-1965 0.39 Alexander and Rowland 1966

'Muscle tissue.

DISCUSSION 65Zn inColumbiaRiverwater with seawater,if conservative (no loss of mass), should produce a 6 S Zn FigureIillustrates the surface salinity patterns off Oregon during 5-15 July 1967. These patterns indicate specific activity in proportion to the salinity, assuming the degree of mixing of Columbia River water with the stable zinc concentration in Columbia River water is seawater, with the river water depressing surface salini- equal to that in seawater. Buffo (1967) reports values of zinc concentrations in offshore Oregon seawater ties below ambient seawater values and yieldingan identifiable plume extending offshore southwest of the similar to those reported by Silker (1964) and Kupp river mouth during the summer months. The 32.50/00 and Kroner (no date),for Columbia River water. salinity isopleth has previously been used to distinguish Weighted averages of the musseltissues from each station are shown as data points in Fig. 3. Note that the between Columbia River water and ambientseawater (Budinger, Coachman, and Barnes 1964). This plume line on Fig. 3 denotes specific activity for water and has been traced at sea by salinity determinationsas well

as by radioactivity measurements (Osterberg, Cutshall. MIXED FRACTION OF SEA WATER and Cronin 1965, Park et al. 1965, Gross, Barnes, and 0% 100% Riel 1965, Frederick 1967). During summer months, [Z"3s.w. -!x"Jc.a.w. upwelling takes place along the coast and may forma barrier betweentheplume andthe coastline. The TN association of 6SZn with the plume as well as with coastal organisms suggests that the Columbia River is the prime contributor to 6SZn on the Pacificcoast (Alexander and Rowland 1966). Mussels collected from Yaquina Head (174 km south of the river mouth) have a higher specific activity than would be predicted by linear mixing of river water with 10 Is 50 56 30 seawater. Figure 3 illustrates a simple linear mixing SALINITY (N«) model for Columbia River water of zero salinity with Fig. 3. Linear mixing model. Effect of mixing Columbia seawater of 330/0o salinity. An approximate mean River water (C.R.W.) having a 6SZn specific activity of 1.0 specific activity (1.0 pCi 6SZn/g Zn) for organisms in ,uCi/g Zn with Pacific Ocean seawater (S.W.) of equal stable zinc the Columbia River estuary is taken as the specific concentration. If mussels had the same 6SZn specific activity as activity of the undiluted Columbia River (unpublished their seawater environment, they should fall on the line. Note that the mussels from Tilktmook Ilcad (TII), Yaquina (lead data). Organisms are assumed to reflect the specific (Yll), and Cape Arago (CA) have higher 65Z_n specific activities activity of the water in which they are found. Mixing of than predicted by the linear mixing model. 106

750

that the specific activity for organisms ought to fall on water and transport the zinc onshore. As an unlikely or below the line. Observed 6SZn specific activities example of the latter, an organism might accumulate exceed thevalues expected from linearmixing. 6 sZn and Znnear theriver mouth and then be This situation might occur if the stable zinc concen- transported to the beach, where it is eaten bymussels. tration in Columbia River water were higher than that of Pacific Ocean water. However, if the difference in SUMMARY observedspecific activity at Tillamook Head from that predictedby the linear mixing of seawater and river We have reported 6SZn specific activities in mussels water of equalzinc concentration is ascribed to actual alongthe Oregon coastline. Measuredvalues areprovoc- inequalityin zinc concentration, the Columbia River atively higher than would be expected. would needzinc concentration approximately two orders of magnitudehigherthan seawater(Cutshall, REFERENCES CITED Renfro, and Larsenin preparation). An error of two Alexander, G. V., andR. H. Rowland.1966. Estimation orders of magnitude in trace metal analysis is thought of zinc-65 backgroundlevelsfor marine coastal to be unlikely. Measurements of zinc in seawater and waters. Nature210: 155-57. Columbia Riverwater by independent investigators Budinger, T. F., L. K. Coachman, and C.A. Barnes. yield very similar concentrations (Goldberg 1963, Rona 1964. ColumbiaRivereffluentinthe northeast et al. 1962. Silker 1964, Kopp and Kroner no date, Pacific Ocean,1961,1962:Selectedaspects of Buffo 1967). Itis therefore clear that deviations of' physical oceanography. Seattle, University of Wash- observedspecificactivitiesfrom thelinear mixing ington, Dept. of Oceanography. 78 pp. Technical model are not entirely caused by error in the assump- Report No. 99. tion that zincconcentrations in seawater are equal. Buffo, Lynn Karen. 1967. Extraction of zinc from Zinc-65 from another source also seems unlikely, seawater. M.S. thesis. Oregon State University, Corval- because specificactivities are greater than those re- lis. 49 pp. ported for "Zn from sources other than the Columbia Burt, Wayne V., and BruceWyatt. 1964. Drift bottle River(Alexander and Rowland 1966, Wolfe 1970). observations of the Davidson current off Oregon, pp. Another possible explanationisthatthe salinity 156-65 in: Studies in Oceanography, K. Yoshida isopleths reflectmean conditions and that lower salinity (Ed.). Tokyo. (i.e., higher Columbia River water content) waters Folsom, T. R., D. R. Young,J. N.Johnson, and K. C. occasionallydo reach the coast of Oregon. Drift bottles Pillai.1963. Manganese-54 and zinc-65 in coastal released offthe Oregon Coast in the summer of 1962 organisms of California. Nature 200: 327-29. indicated that variable onshore surface water transport Frederick, Lawrence Churchill. 1967. Dispersion of the does take place (Burt andWyatt 1964). The plume may Columbia River plume based on radioactivity meas- becomehighly convoluted, exhibiting many eddies, urements. Ph.D. thesis. Oregon State University, Cor- with the easternedgebeing especiallyerratic and vallis. 134 pp. reaching the beaches along the northern Oregon Coast Goldberg, E. D. 1963. The oceans as a chemical system, (Panshin1969). If occasional contact with water high in pp. 3-25 in: The Sea, vol. 2, The Composition of Columbia isresponsible, River water content itis Seawater, M. N. Hill (Ed.). Interscience, New York. necessary that either Gross, M. G., C. A. Barnes, and G. K. Riel. 1965. 1. the river water content be very much higher during Radioactivity of the Columbia Rivereffluent. Science theseperiods, since the exposure time is not long. or 149: 1088. 2. zinc assimilationrates of mussels be higher than Kopp, John F., and RobertC.Kroner.No date (about 1968). A five-year summary of trace metals in rivers averageduring these exposures. and lakes of the United States (Oct. 1, 1962-Sept. Neither oftheseseems especiallylikely, although 30, 1967), in: Trace Metals in Waters of the United available data do not afford a test of either hypothesis. States Department of theInterior. Federal Water Two other possibilities are rioted. Relatively higher Pollution Control Administration. Division of Pollu- specific activity 65 Zn from the Columbia River may he tionSurveillance.Cincinnati. Ohio. Various paging. in a chemical or physical form that is more readily Nagaya. Yutaka, and T. R. Folsom. 1964. Zinc-65 and assimilated bymusselsthanthe form of zincin other fallout nuclides in marine organisms of the seawater. Alternatively, some unknown process may California coast. Journal of RadiationResearch 5: separate Columbia River zinc from Columbia River 82-89. 107

751

National Research Council. 1962. Disposal of low-level nisnts,sediments, and water collected from the radioactive waste into Pacific coastal water, a report coastal and offshore waters of Oregon and Washing- of the Committee on Oceanography. Washington. ton. 1961-1963. University of Washington. Labora- D.C., National Academy of Sciences- National tory of Radiation Biology, Seattle. 73 pp. Research Council, 1962. 87 pp. National Research Silker, W. B. 1964. Variations in elemental concentra- Council, Publication 985. tions in the Columbia River. Limnol. and Oceanogr. Osterberg. CharlesL., Norman Cutshall, and John 9(4): 540-45. Cronin. 1965. Chromium-51 as a radioactive tracer of Toombs,George L., and Peter B. Culter. 1968. Lower Columbia River water at sea. Science 150: 1585-87. Columbia River environmental radiological survey in Paushin, Daniel A. 1969. Oregon StateUniversity Oregon. Comprehensive Final Report 1961-1967. Albacore Central Bulletin69-2. Marine Advisory Division of Sanitation and Engineering. Oregon State Program, Sea Grant. Oregon State University, Dept. Boardof Health. 115 pp. of Oceanography, Corvallis. No paging. Watson, D. G., J. J. Davis, and W. C. Hanson. 1961. Park, Kilho, Marilyn J. George, Yasuo Miyake, Katsuko Zinc-65 in marine organisms along the Oregon and Saruhashi, Yukio Katsauragi, and TerukoKanazawa. Washington coasts. Science 133: 1826. 1965. Strontium 90 and caesium-137 in Columbia Wolfe.DouglasA. 1970. Levels of stable Zn and e s Zn River plume, July 1964. Nature 208: 1084-85. inCrossostrea t'irginicafrom North Carolina. J. Fish. Rona, Elizabeth, Donald W. Hood, LowellMuse, and Res. Bd. Canada 27: 47-57. Benjamin Buglio. 1962. Activation analysis ofmanga- Young, D. R., and T. R. Folsom. 1967. Loss of 65Zn nese and zinc in seawater. Limnology and Ocean- from the California sea mussel Mytilus calijbrnianus. ography 7: 201-6. Biol. Bull. 133: 438-47. Seymour, Allyn H., and Gary B. Lewis. 1964. Radio- nuclidesof Columbia River originin marine orga- 108

reprinted from: RADIONUCLIDESIN ECOSYSTEMS Proceedings of the ThirdNational Symposium on Radioecology,D. J.Nelson,Editor CONF-710501 (1973) RLO-2227-T12-6

SEASONAL RADIONUCLIDE INVENTORIES IN ALDER SLOUGH, AN ECOSYSTEM IN THE COLUMBIA RIVER ESTUARY William C.Renfro Department of Oceanography, Oregon State University

ABSTRACT

Neutron activation products from the Hanford plutonium production reactors are present in all components of the Columbia River Estuary. Total amounts of the most abundant gamma emitters (65Zn, "Cr, and 46Sc) in Alder Slough water, sediments, plants, and animals were periodically estimated during 1968-1970. Inventories were made during winter, spring, and summer. Total amounts of the radionuclides in the ecosystem declined as Hanford reactors were shut down. By far, the greatest amounts of radionuclides were associated with the sediments. The biota accounted for only a small fraction of the total radioactivity present. Large seasonal fluctuations were noted in the total levels of the radionuclides in plants, animals, and detritus.

INTRODUCTION part of a Georgia salt marsh. Both radionuclides were rapidly lost from the water to the sediments. Phos- From 1944 to early 1971 plutonium production phorus-32 apparently remained in the upper layer of reactors atHanford, Washington, added small, but the sediments, and the dominant marsh plant, Spartina measurable, amounts of neutron-induced radionuclides alterniflora,did not take up32P from its subsurface to the Columbia River. Perkins, Nelson, and Haushild roots. Various mollusks. crustaceans, and fishes accum- (1966) measured 46Sc, 5tCr, S4Mn, "Co, 65Zn, ulated both radionuclides, with 65Zn being lost from 95Zr-95Nb, '06Ru, 14Sb, and 140Ba in river water the animals muchmore slowly than32P. However, the and suspended particulatesbymultidimensional majorreservoirfor 6 5 Zn was shownto be the sedi- gamma-ray spectrometry between Hanford and Van- ments. couver, Washington. Farther downstream in the Colum- Parker (1962) studied the distributionof stable Zn in bia River Estuary we routinely measure 46Sc, 51 Cr, and thewater,sediments, and biota of Redfish Bay in 65Zn (and less frequently 54Mn and 60Co) in estuarine Texas. He found that although the fishesand shrimp samples. The presence of these readily detectable contained several thousand times the Zn concentration radionuclides in water, sediments, plants, and animals of the bay water, their biomass was small compared provides an excellent opportunity to study the parti- His inven- tioning of elements in an estuary. This paper evaluates with that of the vegetation and sediments. flat showedthat sediments and the relative importance of a number of ecosystem tory for Znin this grass constitute the majorreservoirs. components as reservoirs of radionuclides. An attempt vegetation is made to determine how each reservoir varies in Cross, Duke, and Willis (1970) measuredthe concen- importance in time and, finally, to briefly consider trations of Mn, Fe, and Zn in the Newport River interactions (transfers) of the components one with Estuary, North Carolina. They observedthat all three another. elementsin the sediments varied with timeand sedi- The distributions of various stable and radioactive ment composition but decreased in a seawarddirection. metals in estuarine ecosystems have been studied by a Concentrations of Mn, Fe, and Zn in sixspecies of number of investigators. Duke, Willis, and Price (1966) polychaete worms suggestedthatthe worms were created their own estuarine communities by adding capable of regulating their trace-metal content. plants, shellfish, and fishesto two concrete-walled Phelps (1967) studied the partitioningof stable Fe, ponds then spiking the water with 65Zn. Within 24 hr Zn, Sc, and Sm in benthic communities of Anasco Bay, most of the 65Z.n was lost from the water to sediments Puerto Rico. He observed that Fe was presentin highest or through tidal exchange. One hundred days after amounts inboth fauna and sedimentsbut that Zn was addition of the 65Zn, 94-9954 of the 6SZn present in concentrated in animals to higher levelsthan in the the experimental ponds was in the top 6 cm of the sediments. He concluded that the density and species sediments. composition of the infauna have a direct influence on Pomeroy. Johannes, Odum, and Roffman (1969) the partitioning of Fe, Zn, Sc, and Smin the com- followed the distributions of 32P and 65Zn spikes munity. For example. Fe was concentrated in poly- added simultaneously to the water of the Duplin River, chaete worms which feed selectively on specific sizes of

738 109

739

PACIFIC.1 OCEAN WATER

Inventory Boundary COLUMBIA RIVER WATER

JMS1

PACIFIC OCEAN

KILO

DEPTHS IN METERS INVENTORY AREA = 6 99O m2 VOLUME AT HIGH TIDE LEVEE (1O.5m) = /7,2OOm3

Fig. 1. Map of the Alder Slough area showing depth contours and boundaries of the inventory area. The inset is an aerial photograph of the Columbia River Estuary from 20,000 m. the most recently settled particulate matter. In con- Threetimes of the year appearedto be most trast,polychaetes which move through the upper important for making inventories. In midsummer the sediment layers feeding less selectively tend to concen- total biomass of all the plants and animals is greatest. trate Zn to the exclusion of Fe. The sedges and algae grow in rank profusion throughout To investigate the behavior and distribution of radio- the intertidal zone. Large numbers of small crustaceans nuclides in all the components of such a large, complex andfishestake advantage of therelatively warm region as the Columbia River Estuary would be a major productive slough waters, and these, in turn, attract undertaking. For this reason, Alder Slough (Fig. 1), a larger carnivores to the area. As winter approaches, the discrete portion of the estuary small enough to ade- solar insolation and water temperatures diminish. By quately sample on a reasonable time scale, was chosen December most of the sedges and algae have died and for the study. This small, "L-shaped" arm of a tidal flat only a few animals remain in the area, so that the total on the southern shore of the estuary has a number of biomass of the biota is minimal. Spring is a transitional desirable features: (a) it is accessible by land or water, season between the barren winter period and the high (b) it is shallow enough to seine, yet deep enough to standing stocks of the summer, so this period represents retain water at all tide levels, and (c) it is a protected an intermediate stage between the two extreme seasons. site of relative calm even during storms. 110

740

METHODS amphipod, Anisogammarus confervicolus, and the isopod, Cnoriniosphaeronta oregonensis. were each arbi- The general approach to estimating total amounts of trarily assumed to be one-tenth thatof C. salmonis. the radionuclides in the arbitrarily delimited inventory The total weightsof fishesand larger crustaceans were area (Fig. 1) involved taking representative composite determined from seine collections. During each inven- samples from measured areas and then multiplying by tory a small-mesh 15-m bag seine was used to capture the total area. A 12.7 X 12.7 cm Nal(TI) well crystal animals from an area of the bottom representing 4.5% and photomultiplier coupled to a5 12-multichannel of the total inventory area. The biomass of each age analyzer was used for all radionuclide analyses. The class of each species taken in the seine was projected to resultant gamma-ray spectra were resolved by a least- the total area, and appropriate radioanalyses were squares computer program. carried out. Sediments were obtainedatten locations in the The total activities of the gamma-emitting radionu- 14N slough by inserting an inverted glass Petri dish into the clides 6s Zn, 51Cr, 46Sc, 60Co, and 1n in each of the mud surface and extracting the niud-filled dish by hand. AlderSlough ecosystem components are presented in The ten samples were thoroughly mixed in a plastic bag the figures and tableswhichfollow. Confidence limits and then returned to the laboratory. Subsamples of the in the form of single standard deviations are shown. In sediment were dried at 60-70 C and packed in plastic all cases the standard deviations are based on errors due counting tubes for radioanalysis. Activities of replicate to counting statistics, weighing, andbiomass extrapola- subsamples from the well-mixed mud samples were in tions. When the single standard deviationexceeded 33% close agreement, so that lack of sample homogeneity of the nominal value, a dashed line was used. In those did not appear to be a problem. cases in which the activity of the radionuclide wasless Detritus was taken from the uppermost1 cm of than 0.01 pCi, a zero was recorded on the inventory. sediments at three points in the inventory area. A circular dip net was pressed I cm into the surface of the RESULTS AND DISCUSSION mud, aqd all the material within the circle was removed and strained through the 0.5-mm meshes of the net. Water. The radioactivityof Alder Sloughwater is The resulting detritus, consisting mainly of decaying influenced by the mixing of waters from three different plant particles, was washed briefly to remove mud, sources: the Columbia River, Alder Creek, and the dried, ashed at 450 C, and radioanalyzed. Pacific Ocean (Fig.1). The mixed, semidiurnal tidal The biomass of green algae in the inventory area was cycle results in two high and two low water levels each estimated by scraping all the algae from a measured area day. Coupled with these tidal exchanges are seasonal of the tide gate wall. The weight of this sample variations in the discharge rates of the river and the multiplied by the total number of equal areas in the creek. The net resultof thevaried inputs of water into slough covered by algae of similar density produced an the inventory area is a constantly changing mixture of estimate of total weight. ocean, creek, and radioactive river waters to produce a The total weights of the intertidal sedges, Carex and continuous change in the concentrations of radionu- Scirpus, were determined by counting the number of clides in the slough water. "average" plants in three randomly selected quadrats On the first inventoryin July1968, the concentra- and estimating the total area covered by the plants on tions of6"Zn,"Cr, and "Sc in both particulate and scale maps. An "average" plant of each species with ionic forms were measured in. surface and near bottom root system intact was washed, dried, ashed, and -waters at 2-hr intervalsfor 24 hr.The concentrations of radioanalyzed. these radionuclides in both ionic and particulate form The biomasses of the small crustaceans which inhabit in the slough water had the following ranges:65Zn. the slough throughout the year are difficult to estimate. 0-16pCi/liter,5 t Cr. 50-246pCi/liter, and "Sc,2-5 Numerous counts of the ubiquitous amphipod, Cor- pCi/liter.If the 24-hr mean activities per liter are ophium salmonis, over several years of study in Alder multiplied by the high tide volume of the inventory Slough suggestedthatthis tube-dwelling animalis area (17.2 X 106 liters), the resultant totals constitute present at a density of about one adult per square substantial fractions of the total activities of the three centimeter. Using this number as the best estimate of major radionuclides: 6 5 Zn, 2.2%o; 51 Cr, 22.0%: 0 6 Sc, population density together with estimates of the 6.070.Except for July 1968, sampling and inventories weight of individual amphipods, the total ash weight of were done within an hour or so of low tide. As a result, C. sabnonis was set at 6990 g. The biomasses of the the slough surface water was dominated by Alder Creek 111

741

water, and radioactivities were low and hence not representativeof the conditions prevailing throughout the day. For this reason, estimates of the radionuclide 8000 a'Cr / totals in water werenot attempted, although the water may be an important reservoir of the system. SEDIMENT Sediment. Figure 2 shows how the radioactivity in (top cm) sedimentsvaried seasonally and over the period of the study. Both 65Zn and 96Sc declined steadily through- 6000 out the 21/2-year period, probably due to shutdown of single-passnuclear reactors at Hanford. In February 1968, Reactor B was shut down, leaving three plutonium production reactors in operation. Figure 2 shows 40001- the subsequenthistory of reactor operations at Han- 1 / ford. Chromium-51 levels in sediment of the inventory area also diminisheddrastically from 1968 to 1970, but in 1 an interesting 5 1 Cr 1 I fashion. In each year the total 2000 droppedsharply from December of the previous year to 1 April, but thenincreased through December. The rapid drop intotalactivity of 51 Cr in sediments from Decemberto April may result from seasonal flooding of 1__ the sloughwith nonradioactive creek water. During the 0 winter rainy season(November through January) the dischargeof fresh water from Alder Creek is at its annualpeak, dominating the hydrography of the slough 200 (Fig. 2), This creek water is deeply colored and appears DETRITUS s'CN (too cm) to be highly concentrated in organics dissolved from the \ peat bogsand alder thickets in its 5 km2 watershed. / The large volumes of this water may displace some of / the radioactivity associated with sediments. Hence, 100 some radionuclidesin the sediments may be lost during / winter creekflooding,to be replaced during the remainder of the year. / i52e A 46Se d" Detritus. Detritus in Alder Slough consists largely of ------o C ----- , brokenand decaying plant material resulting from the 0 die-off of the large biomass of sedges in and around the slough. This detrital material is not highly visible, often being coveredwith sediment, but itis an important REACTORS reservoirof radionuclides and undoubtedly forms the energybase for some of the food webs in theslough. The total amounts of 65Zn and 46Sc associated with L1. detritus ofthe inventory area remained roughly constant during the 2'/2 -year study. 20 RAINFALL As in the sediments,5declined'Cr drastically over 4 the studyperiod but reached peak totals in the detritus e P during December of each year. The totalmass of OA0.0 9/1 detritus in the uppercentimeterof sediment (Table 1) J 0 A J t A J variedwidely and without pattern around 1000 kg. /966 /969 /9707 D Hencethe decline in the total activity of 51 Cr was not merely Fig. 2. Total microcuties of SIC,, 65Zn, and 46Sc in due to changes in the quantities of detritus sediment and detritus within the inventory area. (Below) present, but to seasonal changes in the amounts of 51 Cr Numbers of plutonium production reactors operating at in the detritus. Hanford and monthly rainfall in the Alder Creek Watershed. Table 1. Alder Slough radionuclide inventories (in microcuries)

SEDIMENT bezn SICr "Sc ''Co "Mn ENDEMIC CRUSTACEANS (Detrital Feeders) Species 6sdn SIC, '"Sc "Co e'Mn July 68 29130343 6823±709 1259±152 Dec 68 1973±239 7912±751 7940111 July 60 CAG 11.10±0.27 0 0 Apr 69 ------° 1023±322 2795±770 8441182 Dec68 CAG 0.77±0.13 3.4300.50 02500.06 0.1900.05 0. 1000.03 July 69 10/30/42 3630±475 645± 78 ---- Apr 69 CAG 1.33±0.24 3.48±0.61 0.37±0.10 0 0 Dec ,69 9240 95 46281425 5410 56 116±27 88±19 July 69 CAG 6.52±1.16 0 0 0 1.8300.34 Apr 70 832± 60 884± 84 612* 40 184±19 59±10 Dec 69 CAG 0 0 0 0 0 July 70 622± 66 1139±180 316* 39 98±22 59017 Apr 70 CAL 1.50±0.20 0 0 0 0 Dec 70 5761 81 1865±367 1691 50 95029 0 July 70 CAL ----- 0 0 0 0 Dec 70 CAG 0.0500.001 0 0.01±0.0020.0120.001 0 Dry Vt. ssZn C ... Corophiun salnonis, A Anisogamsarus confervicolus, DETRITUS(kg) 'ICr ''Sc ''Co s'MI$ L ... Gnorieosphaerena oregonensis

July 68 69327.4±2.9 36.7±8.7 5.2±1.0 0 Dec68 1,093 41.715.4 208.5019.8IS 1±2 6 ---- 7.9±2.6 ENDEMIC CARNIVORES Apr 69 1,355 74.0±11.0144.0±24.242.3±5.5 Species ssZn SIC, ''Sc "Co July 69 481 27.1±2.1 34.2±3.8. 6.610.7 1.710.4 1.210.4 Dec69 1,01822.912.4 89.6±7.8 10.611.1 2.610.6 2.210.5 July 68 PCL 0.0410.01 0 0 Apr 70 483 10.2±0.6 9.5±2.0 5.4±0.5 ---- 1.210.2 Dec68 C 0 0 0 0 July 70 1,945 '31.103.8 . 26.305.9 7.011.2 4.500.8 ---- Apr69 PL 0.0300.01 0 0 Dec 70 943 15.0±,1.7 36.8±6.6 4.411.0 1.90.5 2.110.5 July 69 PCICa 0.0500.01 ----- 0 0 Dec69 PCL 0.0120.001 -- 0 0 Apr 70 PCL 0.01±0.001 0- 0 0 0 GREEN Dry Vt. July 70 PCLCa 0.0220.002 0 0 0 0 ALGAE (kg) ssZn ''Cr '"Sc "Co "Mn Dec 70 PC 0 0 0 0 0

July 684.94 0.500.1 2.0±0.3 0.500.1 0.110.030.041:01 P ... Platichthysstellatus, C ... Crsn822franclnc run Dec68 0 0 0 0 0 0 L ... Leptocottus arnatus, Ca ...Cottus aspen Apr690.29 '1.0100.001 0 0.02±0.002 ------July 69 1.53 0.3900.08 1.2900.21. 0.3800.05 ---- 0.09±.02 Dec69 0 0 0- 0 0 0 OCCASIONAL CARNIVORESAND NERDIVOIIES Apr 700.43 0.0100.002 ------0.0110.002 ---- 0 Species Z, 'Cr ''Sc "Co "Mn July 700.77 0.02±0.0040.031.0060.02±0.002 0 0 Dec 70 0 0 0 0 0 0 July 68 MC 0.3510.04 0 0 Dec68 0 0 0 0 0 Apr69 MN 0.2020.03 0 0 Dry Vt. July 69 OMCMS 0.5210.06 0 0 ssZn seCo CARER (kg) "Cr `Sc Dec69 M 0.0110.001 0 0 Apr 70 ON 0 0 0 0 0 July 68 355 101.8±21.9145.7144.1 46.915.5 6.701.7 0 July 70 OMCG 0.0300.004 0 0 0 0 Dec68 542 7.410.9 15.802.4 Dec 70 G 0 0 0 0 0 Apr69 599 1.310.2 1.910.3 0.410.1 0.160.03 July 69 779 3.810. 12.012.3 1.220.4 0 . Oncorhynchustshwytscha, N ... Mylocheilus caurinus Dec69 149 3.010 12.112.0 0.900.3 1.0±0.2 C ... C1enattgaster aggrrgata, N ... Neonysis e.e rcedis Apr 70 683 3.700.4 2.7±0.7 1.500.2 ------0.420.1 P... Pacifastacustnwbridgii, Ms... MicroDterus selmoides July 70 874 3.300.4 4.521.1 1.010.2 0.210.040.12.02 G ... Gasterosteus aculeatus Dec 70 101 0.300.1 0.700.2 ----- 0.1±0.03. ----

a Single Standard Deviation based on errors Incounting,weighing, and Cry Vt. bionass ettiations. ssZn SC IRPUS (kg) "Cr "Sc "Co ''Mn ... Standard Deviation > 33%. July 681,722 53.216.3 207.0123.2 37.904.4 5.501.4 0 Dec 68 0 0 0 0 0 0 Apr 69 95 1.610.2 14.813.2 0.410.1 0.2±0.04 ---- July 69 479 7.901.3 36.714.9 6.900.9 ----- 1.100.3 Dec 69 0 0 0 0 0 0 Apr 70. 0 0 0 0 0 0 July 70 1.845 8.501.5 15.913.6 3.300.7 4.000.7 0.2±0.4 Dec 70 688 2.720.4 4.011.1 ----- 0.900.2 ---- 113

743

600 / 1. I /s/ EXPERIMENT 400 /969-1970

200 _t 1-1 2nd EXPERIMENT .i_22-. /970-1971 4 0

KIsi EXPERIMENT /969-1970

II I Va 40 c N zed EXPERIMENT4'q/I fl e 0 /970-197/ SEP OCT NOV DEC JAN FEB MAR APR

Fir. 3. Radionuclide concentrations in dead plant material held in nylon bags in Alder Slough.

Although "Zn and "Scin thesediments and Plants. The dominantplantsfound in the inventory detritus generallydeclined, 5' Cr increased to peak area are the intertidalsedgesCarex sp..Scirpus sp., and values in December of each year. One hypothesis the greenalga,Enteromorpha intestinalis(sometimes regarding the increase of5 'Cr1 Cr in detritus and sediment mixed with another unidentifiedgreenalga). It) early to a peak in December is that detrital plant matter in springthese plantsbegin a gradualincreasein biomass the sedimentacts as a "big black sponge" to continually which culminates in midsummer with dense stands of reduce ionic hexavalent 51 Cr to the trivalent state and vegetationaround the edges of the slough. The green sequesterittosediments(Cutshall. Johnson. and algaehang like a dense curtain from various surfaces.'In Osterberg1966) and to living or dead organisms (Curl, fall much of the plant material begins to slough off the Cutshall, and Osterberg 1965). mainplant, and by midwinterlittle algaeorScirpus To test the "big black sponge" hypothesis, large remain, although the decaying roots and stubble of quantitiesof livingsedges fromtheslough were Carexpersistinto the following spring. The algal chopped into small pieces and secured in bags made biomass (Table I) is low in April, highest in July, and from0.4-mm-mesh nylon plankton netting. The bags negligible in December.CarexandScirpusbiomasses were weightedwith stones and placed on the bottom of were alsogenerally highest in July, lowest in December, the slough to be covered by a thin layer of sediment. and ofintermediateweight in April. Monthly samples of the incipient detritus were then Although the intertidal greenalgae greatlyconcen- obtainedand radioanalyzed. Figure 3 shows the result trate many radionuclides from the water, their biomass of detritus bag experiments carried out over two winter in the slough inventory areawas not greatenough to periods. Zinc-65 concentrations in the plant material constitute alarge reservoir (TableI ). However, the remainedconstant, but51 Cr declinedsharplyin amounts of Carex and Scirpus were seasonally high, and December and January of both winters. Since the significant amounts of radioruclides were tied up with hypothesis required a continual reduction of dissolved theseplants,although the radionuclides were not so s 'Cr(VI)to 5' Cr(Ilt), followed by sorption from water greatly concentrated as in algae. Total amounts of all todetritus, the concentrations of 51 Cr in the decaying the radionuclides declined somewhat during the study plant material should have continually increased. In- period, although the effect of decline is accentuated by stead, the activity of5per'Cr unit. weight declined the initially high total activities for plants in the July rapidlyafterNovember of eachyear.Hence the 1968 estimates. hypothesis is rejected. However, these experiments do Animals. In general, the animals inhabiting the inven- confirm the sharp decline in5 1 Cr associated with tory area are small species or the young stages of larger detritus in the inventory area. species. One group of organisms, the small detrital-feed- 114

744

Table 2. Summary of Alder Slough inventories

Animals Sediment Detritus Plants Animals Total Sediment Detritus Plants u Ci u Ci V Ci u Ci u Cl It It It "Zn

d` July 68 2913 27.4 155.5 1.5 3097 94.1 0.9 5.0 0 Dec 68 1973 41.7 7.4 0.8 202) 97.5 2.1 0.4 0.3 0.2 Apr 69 1023 74.0 2.9 1.6 1102 92.8 6.7 2.6 1.1 0.7 July 69 1013 27. l. 12.1 7.1 1059 95.6 0 Dec 69 924 22.9 3.0 0 950 97.3 2.4 0.3 . 0.4 0.2 Apr 70 832 10.2 3.7 1.5 847 98.2 1.2 1.3 0 July 70 622 31.1 8.8 0.1 662 94.0 4.7 0 Dec 70 576 15.0 3.0 0 594 97.0 2.5 0.5

"Cr 0 July68 6823 36.7 354.7 0 7214 94.5 0.5 5.0 0 208.5 15.8 3.4 8140 97.2 2.6 0.2 Dec 68 7912 0.1 Apr 69 2795 144.0 16.7 3.5 2959 94.4 4.9 0.6 1.1 0 July 69 3630 34.2 40.0 0 3704 98.0 0.9 Dec 69 4628 89.6 12.1 0 4730 97.8 1.9 0.3 0 0.3 0 Apr 70 884 9.5 2.7 0 896 98.6 1.1 1.7 0 July 70 1139 26.3 20.4 0 1186 96.1 2.2 Dec 70 1865 36.8 4.7 0 1906 97.8 1.9 0.3 0

"Sc

July 68 1254 0 85.3 0 1344 93.7 0 6.3 0 0 Dec 68 794 15.1 0 0.3 809 98.1 1.9 0 0 Apr 69 844 42.3 0.8 0.4 BOB 95.1 4.8 0.1 July 69 645 6.6 8.4 0 660 97.7 1.0 1.3 0 0 Dcc 69 541 10.6 0.9 0 552 98.0 1.8 0.2 0 Apr 70 612 5.4 1.5 0 619 98.9 0.9 0.2 1.3 July 70 316 7.0 4.3 0 327 96.6 2.1 0 0 Dec 70 168 4.4 0 0 172 97.4 2.6 0

aeco 1 July 68 0 5.2 12.5 0 17.7 0 29.3 70.7 0 Dec 68 0 0 0 0.2 0.2 0 0 0 100 Apr 69 0 0 0.3 0 0.3 0 0 100 0 July 69 0 1.7 0 0 1.7 0 100 0 0 Dec 69 115.9 2.6 1.0 0 119.5 96.9 2.2 0.9 0 Apr 70 184.0 0 0 0 184.0 100 0 0 0 July70 98.3 4.5 4.2 0 107.0 91.8 4.2 4.0 0 Dec 70 95.0 1.9 1.0 0 97.9 97.0 1.8 1.2

a`nn

July 68 0 0 0 0 0 0 0 0 0 Dec 68 0 7.9 0 0.1 8.0 0 98.8 0 1.2 Apr 69 0 0 0 0 0 0 0 0 0 July 69 0 1.2 1.2 1.8 4.2 0 28.6 28.6 42.8 Dec 69 88.3 2.2 0 0 90.5 97.6 2.4 0 0 Apr 70 58.9 1.2 0.4 0 60.5 97.3 2.0 0.7 0 July 70 58.8 0 0.4 0 59.2 99.3 0 0.7 0 Dec70 0 2.1 0 0 2.1 0 100 0 0

e Total <0.1 DCi. 115

745 ing crustaceans, are present in farily constant numbers activityin the areaisconservatively estimated by throughoutthe year. These-include the tube-building limiting the inventory to the top centimeter. During the amphipod,Corophium salmonis:the gamma rid amphi- 2'/2-year period of this study the total activity of6 5 Zn pod, Anisogananarus conferricolus: and the small iso- declined fivefold (Table 2), primarily due to reductions pod,Gnorintosphaeronta oregonensis.These small om- in the levels of "Zn in the sediments. Total levels of nivores constitute a major part of the diets of the other 65Zn in detritus, plants, and animals did not show such animals inthe slough. They are a vital link in the large declines. transfer of energy from detritus up the food webs. 5' Cr. The total amounts of 5 r Cr decreased almost Except for a few instances, when small amounts of fourfold during the study, with abrupt declines after otherradionuclides were measured in (or on) these December each year in the sediment and detritus. In animals, only 65Zn was present in significant amounts. contrast, the activity of 51 Cr associated with plants was The total activitiesof 65Zn in the endemiccrustaceans highestinJuly. Animals did not appear to be a decreasedsharply over the course of the inventories, significant reservoir of5' Cr. with seasonalhighs occurring in July and lows in. "Sc. An eightfolddecrease was observed in the total December as the resultof changes in activityconcentra- activities of "Sc in the inventory area. Most of the loss tions. was due to sharp drops in 46Se levels in the sediments. Carnivorousfishes andsandshrimp which inhabited Scandium-46 totals in the detritus and plants exhibited the sloughduring mostseasons includejuvenile starry no regular trends during the study. flounder,Platichtlfvs ste/laws:young prickly sculpin, 60Co and "Mn. Both 60Co and "Mn became more Coitus asper:juvenile sand shrimp,Crangon francis- apparent in the gamma-ray spectra in the latter part of eorum; and all ageclasses of Pacific staghorn sculpin, the study when thelevels of the more abundant Leptocottus armatus.Without xc tion, only small radionuclides diminished. These radionuclides also were amounts of"Zn were associated with these organisms. predominantly associated with sediments. A number offishes and crustaceans livein the inventory areafor only a portion of the year. These CONCLUSIONS animals, a mixture of both carnivores and herbivores, The percentages in Table 2 show that almost all the include theChinook salmon.Oncorhynchusis/raxy- radioactivity in Alder Slough was contained in the thepeamouth, ,tlvlocheilus caurinus:the shiner tscha; sediments.With fewexceptions,more than 95% of the perch,Cytnatogaster aggregara:the opossum shrimp, total activity of each radionuclide was associated with Neomysis merccdis:the crayfish,Pacifastacus trow- sediments regardless of season. It is clear, then, that for bridgii;the largemouthbass..tlicroptents salnoides:and these radionuclides and others which behave similarily, thethree-spined stickleback.Gasterosteus aculeatus.As we must consider the sediments carefully in attempting was thecase with the endemic animals of theslough, to understand the final disposition of radioactivity these occasionalvisitors contained only small total released to estuaries. amounts of 6 5 Zn. Waldichuk (1961) described sedimentation as". .. ab- Table 2 summarizes the estimatesfor the total straction of dissolved and particulate material from activitiesof each radionuclide in the major components seawater and its disposition on the sea bottom." He of theecosystem. Had the numbers of plutonium stated that sedimentation is a concentrating process in production reactorsat Hanford remained constant over opposition to the desired processes of dilution and the studyperiod, it would be reasonable to attribute dispersion. Hence,the physicochemical characteristics fluctuationsfrom one inventory to the nextto seasonal of estuarine contaminants which may lead to their changes. However, each year began with the shutdown sedimentation should be carefully considered before of onereactor (and, presumably, a corresponding theirrelease. Thisis true whether the contaminants are reductionin the input of radionuclides), so that the radionuclides, toxic metals, or harmful organic com- continually responding to the reduced reservoirs were pounds. inputs. The sections which follow consider each radionuclide in turn. ACKNOWLEDGMENTS 65Zn. Most of the 65Zn in the inventoryarea is in the top centimeter of the sediments. Radioanalyses of core Much of the field work involved in this project was samples(unpublished data) indicate that lower, but accomplished by the careful and diligent effort of measurable, quantities of "Zn are also found in the Norman Farrow. I am indebted to him and to 1. Lauren secondand third centimeter layers. Thus the total Larsen, who radioanalyzed the samples. Thanks also go 116

746

to Drs. Normal Cutshall and William G. Pearcy for zincin experimental ponds.ChesapeakeScience review of the manuscript and constructive comments. 7(l): 1-10. This project was supported by USAEC Contract Parker, P.L. 1962. Zinc in a Texas bay. Publ. Inst. Mar. AT(45-1)2227 Task Agreement 12 (Manuscript No. Sci. 8: 75-79. RLO-2227-T 12-6). Perkins, R. W., J. L. Nelson, and W. L. Haushild. 1966. Behaviorandtransport of radionuclides inthe LITERATURE CITED Columbia River between Hanford and Vancouver, Washington. Limnol. Oceanogr: 11(2): 235-48. Cross, F. A., T. W. Duke, and J. N. Willis. 1970. Phelps, D. K. 1967.Partitioningof the stable elements Biogeochemistry of trace elements in a coastal plain Fe, Zn, Sc, and Sm within a benthic community, estuary: Distribution of manganese, iron, and zinc in Anasco Bay, Puerto Rico, pp. 721-34 in: Radio- sediments, water, and polychaetous worms. Chesa- ecological Concentration Processes, B. A. Aberg and peake Science 11(4): 221-34. F. P. Hungate (Eds.). Pergamon, New York. Curl, H., N. Cutshall, and C. Osterberg. 1965. Uptake of Pomeroy,L. R., R. E. Johannes, E. P. Odum, and B. chromium(111) by particlesinseawater. Nature Roffman. 1969. The phosphorus and zinc cycles and 205(4968): 275-76. productivity of asaltmarsh, pp. 412-19 in: Cutshall, N_, V. Johnson, and C. Osterberg. 1966. Symposium on Radioecology. D. J. Nelson and F. C. Chromium-51 inseawater:Chemistry. Science Evans (Eds.). USAEC CONF-670503. 152(3719): 202-3. Waldichuk, M.1961. Sedimentation of radioactive Duke, T. W., J. N. Willis, and T. J. Price. 1966. Cycling wastesin thesea. Fish. Res. Bd. Canada, Circular of trace elements in the estuarine environment. No. 59: 1-24. 1. Movement and distribution of zinc-65 and stable 117

reprinted front: RADIONUCLIDES IN ECOSYSTEMS Proceedings of the ThirdNational Symposium on Radioecology, D. J. Nelson, Editor CONF-710501 (1973)

RLO-2227-T12-7

RATE OF ZINC UPTAKE BY DOVER SOLE IN THE NORTHEAST PACIFIC OCEAN: PRELIMINARY MODEL AND ANALYSIS HenryA. Vandcrploeg Department of Oceanography Oregon State University, Corvallis

ABSTRACT

This paper presents a specific-activity (SA) model that describes the relationship between the SA of a fish (or any predator) and the SA of its prey. The model serves as the basis for the hypothesis that lowered feeding intensity is the major factor causing the decreased excretion rate constant observed with increasing body weight for zinc and some other elements. Furthermore, the model facilitates understanding of SA differences observed among different species. The model is applied to a free-living population of Dover sole to calculate the zinc uptake constant for the population. The value obtained for the populationis lower than the values obtained from the laboratory retention studies of other workers. The feeding-intensity hypothesis "predicts" these differences.

INTRODUCTION Dover sole (dficrostomus pacificus) to determine the populationuptakeconstant. The Hanford reactors, until just recently, produced a number of radionuclides that entered the Columbia METHODS River and eventually reached the surface waters of thenortheastPacificOcean. One oftheradio- Dover sole were collected at NH-23, a 200-m-deep nuclides, 65Zn, was produced in large enough quan- station 23 nm west of Newport, Oregon. The fish tity, has a long enough half-life, and is sufficiently werecollected ina 7-m semiballoon shrimp trawl concentrated by the biota to appear in measureable with a 0.5-cm-mesh liner in the cod end to prevent concentration in the biota of the northeast Pacific contactof thefish with the ship's deck. Samples Ocean. This study attempts to evaluate the impor- were immediately frozen and were kept frozen until tantfactors affecting 65Zn specificactivities (pCi analysis ashore. Digestive tracts were sampled to pro- 6sZn/g totalzinc)of benthicfishes on Oregon's vide "prey" of the Dover sole. Gut contents from continentalshelf. To accomplishthisgoal ady- different sections of their very long digestive tracts namical model was derived. This model provides a were pooled from allfish collectedon a particular method of quantifying the importantfactors that samplingdate.Thefishandtheirgut-content affectthespecificactivities (SA) of thedifferent samples were each driedat 65 C for two weeks, species.It can also be used to estimate the rate of ashed at 420 C for 72 hr, ground with mortar and uptake of a particular element by a predator from pestle, and packed into13-cc counting tubesfor its prey. gamma-ray counting. A portion of the ash was re- The purpose of this paper is to present the model tained for stable Zn measurement by atomic absorp- and discuss its three principal uses: tion spectrophotometry. Radioanalyses were done of' a 12.7 X 12.7 cm Nal(Tl) well crystal coupled to a (1)to provide a new explanation for the decrease in 512-channelanalyzer. Eachfishand gut content the excretion rate constant of assimilated zinc sample was counted usually for four 400-min periods and some other elements with increasing size of to get desired precision of about ±5%. The counts the organism, , were correctedforphysical decay to the date of (2) to identify important proximate factors affecting collection. The portion of ash wasdigested in con- SA differences among species that accumulate centrated HNO3, then diluted with 0.36 N HCl. and the radionuclide and stable element from their analyzed by atomic absorption spectropltotometry. food, and (3) to allow calculation of food-chain uptake rates THEORY of different elements. Basic model. Foster (1959) in his specific-activity The model is applied to a free-living population of approachtounderstandingradionuclidedynamics

840 118

841

noted the utility of SA in comparing different com- Defining the uptake "constant" a by partments,forexample,sedimentandorganisms havingdifferentstable-elementconcentrations. He r=aZ (5) wrotea differential equation to describe the change of SA in an organismand emphasized itsapplica- and from placing this relation for r in Eq. (4), bilityfor determiningthebiological turnover constant: dX X dZ = aZF(t) -XaZ + XA. (6) dt Z Z dt dS/dt= pC - (), + p)S , (1) We want a specific-activity relationship, thatis, an where expressionfor d(X/Z)/dt. By the quotient rule, C = specific activity of the organism's source of radio. activity, d(X/Z) 1 dX X dZ (7) S =specific activity of the organism, dt Z dt dt

A = physical decay constant, Dividing expression(6)by Zand using this result for S =biologicalturnover or excretionconstant. (1/Z)(dX/dt) in (7), Foster's equationmotivated the development of the SA equationderived here. (8) The equation for changein the amount of 6 s Zn in a fish acting as a single compartment and obtain- (X/Z) by S for specific activity of the fish, ing all 6 5 Zn fromits food is Replacing dS/dt = a( F(t) - S) - SA . (9) dX/dt = r F(t) - r' (X/Z) - XX, (2) If the derivation had proceeded using 0 instead of a, a where more complexrelation thanEq. (9) involving terms X = amount of 6 s Zn in thebody of the fish, containingdZ/dt would have been obtained. By using r = rate of inputof totalzinc into thefish the uptakeconstanta l(zinc input)/(body burden)/ (actually assimilated), (unittime)] , a simpler relation is obtained. If dZ'dt = 0, thatis, no "growth," then or = Q: and Eq. (9) is F(t)=specific activity of prey (pCi 6 sZn/g total equivalent toFosters equation. From Eq. (3) and (5) it zinc), a function of time, can be shown that a= 0 + (I /Z)(dZ/dt). For rapidly r' = rate of excretionof total zinc from fish growing organismsthe term (1/Z)(dZ!dt) could be (pg/day), significant relative to 0(as will be seen below). A = physical decayconstant(0.00283/day). An interpretationof a. An examination and discussion of a arenecessary for two reasons. First, I shall Justificationfor food-chain uptake comes from the hypothesize a new mechanism for the observed relation- experimentsof Hoss(1964) andBaptist and Lewis ship between body weight and turnover time (Eber- (1969). The material that goesin must be conserved, hardt and Nakatani 1908, Golley et al. 1965). Second, so the hypothetical relationship will predict the magnitude of the effect of body weight on the excretion constant. r=r'+dZ/dt, (3) which will then be used to assess the danger of lumping some of the fish of differentsizes and also allow where Z = total zinc in fish.(Note: 65Zn decays to comparison of theresulting a with the O's from stable Cu,but the amountof totalzinc lost in this experimentsperformed with smaller fishes. The bal- way isnegligible.) By substituting forr' from Eq. anced equation of Winberg (1956)relatesration, (3) into Eq. (2), growth, andmetabolism in fishes:

pR=T+AW/At, (10) dX = r F(t) - i (r - d! - XA. (4) 119

842 where If AW/.fit is small compared with AW°6 (as will be R = food intake per unit of time, justified later for some of the fish in the experiment), Eq. (14) may be simplified to T = total metabolism, W = body weight, a=C(AW-0.3), (15) p = constant for correction of ingested to utilizable energy of food. where C is a constant. If the assumptions leading up to Eq. (15) are correct, then a and concomitantly (1 (if Also consider Winberg's relationship between respira- AW/At is small) will follow this weight relationship. tion and weight: According to the assumptions that lead to this interpre. tation, the rate of food consumption is the factor most T=AW3, (11) directly influencing the zinc uptake constant, a. If the zinc binding sites are saturated at a given input rate and where A =aconstant, a function of temperature, and y there is no growth, then it follows that the excretion = 0.8. Assume that rate constant will follow this pattern. Mishima and Odum (1963) observed that the excretion rate constant r=abR, (12) for the assimilated pool of 65Zn in the snail Littorina irroratacould be correlated with temperatureand size. where a is a constant for the assimilation efficiency of They tentatively concluded that the excretion rate zinc,and b isa constant that adjusts for different reflects the activity of the organism. Edwards (1967), concentrationsof zinc in the fish and its food. Then working with young plaice. showed thatthe zinc excretionconstant he observed for the early phase of r ab(AW°6 + AW/At) 65Znexcretion could be correlated with respiration. lie (13) Z= pZ also foundthat excretion was independent of the feedinglevel, although the levels were not defined In Fig. 1, Z is plotted against body weight, W. Placing quantitatively. It should be noted that this correlation the expression for Z in Eq. (13) results in appliedonly to roughly101/"tof the assimilated zinc. Pequegnat et al. (1969) estimated that zinc is accu- ab(AW°.a + AW/fit) mulated inconcentrations many timesthe marine orga- a= p(4.83 W'.oa) (14) nism's needforitin enzymes and suggested that adsorption-exchange is the main mechanism of concentration. This lends support to my contention. Since only a small amountof zinc is bound to enzymes, only a smallfraction can be expected to be influenced directly by metabolic rate. Giventhat organisms are really adsorption exchange systems, feeding intensity could possibly be the major factor causing the observed size-turnover relationship, if theassimilation constant is indeed a constant. The problem of multiple compartments. Often radio- ecologistshypothesize that organisms excrete radionuclides not as if they were single compartments but rather a sumof a number of compartmentsor groups of organs(Eberhardt and Nakatani 1969). ti Perhaps the best evidence formultiple compartment excretionof Zinc in fish comes from Nakatani's (1966) long-term uptakeand retention study of 65Zn in trout. E 300 600 900 1200 He noted aninitial fast decline of 65Zn, possibly the FISH WEIGHT tgJ sum of several fast components. that lasted for a period Fig. I. Relationship between total zinc in a fish and its body of five weeks followed b1 a later slower decline. Since weight. Z = 4.838711'1 .0795: the standard error of the exponent we usually analyze whole fish to determine SA, and :0.0403. since we mustusually work with a nonzeroand often 120

843

,,,,,,constant F(t) as well as a nonzero starting value for 3. The specific activity available to the fish in the food ,;-on Lq. (9), we are forced to work with a single is actually equalto the specific activitymeasured for .tnpartmcnt a or p, that is, a composite a or 0 for a the food as a whole. possibly multicompartntent system. We must ask what Lumping fish of different sizes could potentially lead the a we estimate really represents. to error (Figs.I and 2), because the fish spanned one Consider a for a two-compartment system. Let the order of magnitude (from 96 to 1049 g) in weight. of radioactive zinc inthefirst and second amount Before we can predict the hypothetical importance of ,onipartments be denoted respectively as X, and X2. this range, the importance of the AW/At term relative the change in radioactivity for the whole organism is to AW°'a in Eq (14) must be determined forMicro- the sum of the changes in X, and Xz ; that is, stonnrs.This can be approximated by utilizing the metabolismand feeding relationships established by dX dX, +dXz (16) Winberg (1956), andlatercorroborated by Mann dt dt dt (1967). From Winberg's equation Q = 0.3W°-8, the oxygen consumption of a fish weighing 100 g in a Writing expressions for dX,/dt and dXz/dt analogous closed vessel would be 11.94 ml/hr or 287 nil/day at 20 todX/dt in Eq. (9) and substituting them into Eq. (16) C. gives Microsto,nuswas caught at a depth of 200 in, where thetemperatureremains nearlyconstantat8 C. dZ, X, t= (r, + rz) F(t) - Correctingfor temperature by "Krogh's normal curve" (r,-at)z, gives an expenditureof S2.3 ml of oxygen equivalent to 82.3 mgof dry material per day (Winberg 1956). dZz l Xz l (( + (17)Multiplying this value by 2 to account for the fish's + z extra activityin nature gives the expenditure of 165 mg/day(1Vinberg 1956, Mann 1967). From the data of Substitutingthe right-hand side of Eq. (9) for dX/dt, Hagerman(1952)therate of growth of a100-g substitutingaZ, a, Z, , az Z2 for r,r, , r2 , and solving (yearling) Dover sole is roughly 0.33 g fresh wt or 50 for a leads to mg dry wt per day. Thus _1W/At is about 23% of the sum (AW°-a + AW/At). For the 200-g fish (age 2+ a, X, + azXz years) AW/At is less than 15 G of the sum, and this a X +J(Ldt)@ percentage decreases rapidly with size, so that for fish above 200 g we can ignorethe AWV/At factor in

dZz\Xz/Xj 1 (18) predictingthe effect of weight on a. `dtl`Z, dt Zz/J Equation (I5) predicts that an order of magnitude [(dz,vx'/x\ increase in a fish's weight would cause a to he halved, In Eq.(18) the expression inbracesisprobably even if the growth term is ignored. Also, if the larger insignificantfor most cases, as it represents the differ- fish were rive times the weight of the smaller, it would ence betweentwo small, nearly identical quantities. have an a 61% of the smaller. It would appearthen that Thusa is very nearly a weighted average of a, and a2, size as wellas growth effects could be important as Eber- wherethe weighting factors are the amounts of radio- hardt and Nakatani (1968, 1969) believed. If adecreases activityin each compartment. with size,we should expect that SA would decrease with size under equilibrium conditions. Figure 2 shows SA plotted against body weight for each month. For THE EXPERIMENT March andApril there appears to be a decrease in SA Furtherassumptions. Before we can actually use Eq. with increasing size. However, this trend is not seen for (9)for determining a, more assumptions must be later months, andif the fishless than 200 g are examined; these arc` excluded, the trend is not apparent. What is apparent, however, is the great scatter in some of the SA data that 1. Lumping of the different sized fish in the experi- could mask any SA-weight trends.It should also he ment will not markedly affect the estimate of a. notedthat a SA-weightrelationship may not be 2. The fish do not immigrate from great distances, observed for fish that are not in equilibrium with the particularly with reference to depth after sampling SA of their food. Evidence for lack of equilibrium will has begun. be given below. 9 March 70 /2 April 70 it May 70 27 May 70 16

40 . S.

20 July 70 18 and 26 August 70 /6 September 70 /0 October 70 Q 16r

e ti 4

0 300 600 900 0 300 600 900 0 300 600 900 0 300 600 900 1049 FISHWEIGHT (g)

Fig. 2. Relationship between specific activity and body size at NH-23 for different collections in 1970. 122

845

If, after the initiation of the experiment,Table I. Specific fish activity immi- of food in different sections of gut grate to station NH-23 froth a distant place, they could 65Zn bias theresults, assuming the SA of their prey - No* of fish Date Section of gut specific invertebrate infauna, mostly polychaetes - had been analyzed activity different. Carey (1969) noted a marked decrease in SA of echinoderms with increasing depth (as much as 251 12 April 1970 II Foregutt 0.0235 per 100 m) and a gradual decrease in SA with increasing 12 April 11 Hind gut2 0.0254 distance south of the Columbia River. Dover sole are 10 May 9 Foregut 0.0170 thought tobe local populations which migrate shoreward 10 May 9 hind gut 0.0234 during early spring and offshore during October and No- 27 May 8 Stomach and first 1/33 0.0160 vember (Hagerman 1952). Although most Dover sole tend 27 May 8 Second t/3 0.0181 to remain in asingle locality, tagging studies indicate a 27 May 8 Last t/3 0.0118 few individualsmay move as far as 48-580 km (Westr- 20 July 8 Stomach 0.0222 heim andMorgan 1963). During the winter the mature 20 July 8 First 1/3 0.0213 0.0206 Doversole usually spawn at depths greater than 250 m, 20 July 8 Second t/3 20 July 8 Last '/3 0.0210 whereas juvenilesand some immatures (mostly smaller 18 August 5 Stomach and first t/3 0.0193 fishthan usedforthisstudy)remain atdepths 18 August 5 Second and last f/3 0.0215 shallower than200 m (R. Demory, personal communi- Stomach and first t/3 0.0199 cation). For this experiment itis assumed that all the 26 August 8 26 August 8 Second and last t/3 0.0168 fish originatedfrom the same depth, migrated to the 200-m station before initiation of the experiment in 16 September 11 Stomach and first t/3 0.0179 16 September 11 Second and last f/3 0.0190 March,and thatother fish did not immigrate into the 0.0179 study areaduring the experimental period March 10 October to Stomach and first t/3 10 October 10 Second and last f/3 0.0196 through October. The possibility that fish might absorb a different SA !Stomach contents plus first one-fourth of intestine. from their food than measured for the food as a whole 2Last three-fourths of intestine. was thoughtto be a potential problem. For example, 3First t/3 is first one-third of intestine. less easilydigestibleparts of the prey (such as a clamshell). could have a different SA from the softer matesof a and S(0) are used to generate a solution for parts, havinga higher turnoverrate.To test for this S. Eachestimateis changed by the program until an a problem andto determine whether pooling of the gut and an S(0) are obtained that minimize the sum of contentsfrom the entire length of the alimentary tract squares around S andthedata points for some biased thecalculations, I divided the gut into a number convergence criterion. of sectionsand determined specific activity for each To simplify that mathematics, F(t) was assumed to be sectiontoseeif SA of the contents increased or a constant, the "true" mean F of F(t); that is, decreasedwith distance away from the stomach. Table ftf I shows thatno consistent trend appeared. F(t) dt tl Uptake rate. Figure3 shows the SA of the fish and F= the pooledSA of their prey, thatis, the weighted tf - ti average of all gutcontents sampled on a particular date, where F(t) is the function represented by the dashed wherethe weighting factors are the amounts of total lines connecting the values for specific activity of the zinc per sample. food in Fig. 3, and t1 and tf are initial and final times Data fromMarch through October 1970 were used to respectively. Then the solution to Eq. (9) is calculate or, The September 1969 data are included to showthat the fish probably are not in equilibrium with their food during theexperimental period. This would (S(O) _ .__ ) e -(a+ a)t rationalizethe decrease in SA of the fish with time S(t)a+ X+ during this period. To determine the best a for a given F(t), an analytical The nonlinear least-squaresprogram*CURVFITI was solutionof Eq. (9) is helpful. Initial estimates of a and used for calculation of a. For the first analysis, all fish of the initial condition on S, thatis, S(0), can be utilized by aleast-squares-minimization program. Esti- 1. Available from Oregon State University Computer Center. 123

846

34 0

` 24 10 b True mean \ 0_____O S

O, 18

I a I 12 Best fit o// fish

I 6 Best fit fish,-200gc I

0 SEP FEB MAR APR MAY JUL JUL AUG OCT /969 /970 Sing/e fish . 200g ISingle fish r 200g o Pooled gut contents

Fig. 3. Specific activity of the Dover sole and its prey at NIt23 during the sampling period.

(except the two exhibiting extremely low values in Fig. with a standard error of 2.845 X 10-4. Considering the 2) were included. Figure 2 shows that fish less than 200 standard error on a, the approximation of F(t) seems g had much sigher specific activities than the other fish. reasonable, although further analyses using a more This could be owing to the size and growth rate exact representation of F(t) will be tried. phenomena already discussed or to the possibility that in the fall before the experiment the smaller fish did DISCUSSION not migrate into deeper water (where the SA is lower) with thelargerfish. The higher SA's could be a Assuming a = 0, we cast tentatively compare the re- combination of allthree factors. For these reasons sults of this experiment with others. Table 2 compares another determination of a was made for the fish the values of P. calculated from effective half-lives given greater than 200 g. The best least-squares fit for these by other authors. with the a calculated for Mlicrostotnus two cases is shown in Fig. 3. Another regression was in this study. The value for a is lower than the values computed for the fish less than 200 g. Then, the sum of for 0 of the other experiments, although it is not far squares (SSE) about the regression of all the fish was from the lowest value of Renfro and Osterberg (1969). partitioned into two subsets: the first, the sum of the A comparison of the a ofllicrostonuts with the 0's of SSE for the fish >200 g and the SSE for the fish <200 the small flounders of Renfro and Osterberg can be g; and the second, the difference between the SSE of all made, assuming size and temperature are the factors fish and the first subset. causing these differences. Krogh's curve predicts food An F test showed that the mean square of the consumption of fish at 15 C would be 2.2 times higher difference subset was significantly greater than the than at R C. The size difference between Microstoruus mean square of the first subset SSE at the I7 level. and the little flounders (about two orders of magni- Therefore, the a for the fish >200 g is reported. The tude) predicts that the a ofAlicrostonats should be 25% value of a calculated for fish >200 g is 1.0922 X 10-3, of that of the smaller fish. Taking both size and temper- 124

847

Table 2. Comparison with other results

Corrected a or p Author Species Size Temperature (C) Comments a

0.00954 0.0097 Renfrn and Platicluhys stellatus 0.3 g dry wt Is Fish emaciated, O's Osterberg (1969) are highest and lowest of 3 individual tisti 0.00145 0.0097 0.00212 Rice(1963: Paralichthyssp. Postlarval Calculated by Renfro and Osterberg,1969) 0.00234 0.0038 Nakatani(1966) Salrno gairdneri 100 g at start 11-22 Chronic dose 1000 g at finish

Nakatani (1966) Sahnogairdneri 336 g 20 Single oral dose 0.002t2 0.0038

This study Microstomus pacificus200-1049 g 8 Free-living population 0.00109 0.00109

ature into considerationleads to the conclusion that ACKNOWLEDGMENTS Alicrostornusof a size (ignoring growth) comparable to The data,results, and conclusions of this paper are Renfro's floundersand maintained at similar temperatures wouldhave an a about0.0097. slightlylarger than part of a doctoral dissertation in progress. under the direction of Dr. William G. Pearcy at Oregon State the largest Qhe observed. This correctedaand others I am especially grateful to Dr. Pearcy for arc shown in the last column of Table 2. University. both his scientific guidance and personal encourage- ForNakatani's trout the temperature factor of 3.5 between 8 and20 C implies that theafor

848

Eberhardt,L.L., and R.E. Nakatani.1968. A Mishima, J., and E. P. Odum. 1963. Excretion rate of postulated effect of growth on retention time of zinc-65 by Litiorina irrurata in relation to tempera. metabolites. J. Fish. Res. Bd. Canada 25(3): 591-96. ture and body size. Limnol. and Oceanogr. 8(l): Eberhardt, L. L., and R. E. Nakatani. 1969. Modeling 31--38. the behavior of some radionuclides in some natural Nakatani, R. E. 1966. Biological response of rainbow systems,pp. 87--109 in:Symposium on Radio- trout (,Salnto gairdncri) ingesting zinc-65, pp. 809-23 ecology, D. J. Nelson and F. C. Evans (Eds.). USAEC in: Disposal of Radioactive Wastes into Seas, Oceans CONF-670503. CFSTI, Springfield, Va. and Surface Waters. IAEA, Vienna. Edwards, R. R. C. 1967. Estimation of the respiratory Pequegnat. J. E., S. W. Fowler, and L. F. Small. 1969. rate of young plaice (Plcuronectes platessa L.) in Estimates of zinc requirements in marine organisms. natural conditions using zinc-65. Nature 216(5122): J. Fish. Res. Bd. Canada 26: 145-50. 1335-37. Renfro, W. C., and C. Osterberg. 1969. Radiozinc Foster, R. F. 1959. Radioactive tracing of the move- decline in Starry flounders after temporary shutdown ment of an essential element through an aquatic of Hanford reactors, pp. 372-78 in: Symposium on community with specificreferenceto radiophos- Radioecology, D. J. Nelson and F. C. Evans (Eds.). phorus. Pub. Staz. Zool. Nap. 31: 34-69. USAEC CONF-670503. CFSTI, Springfield, Va. Golley. F. B., R. G. Wiegert, and R. W. Walter. 1965. Rice, T. R. 1963. Review of zinc in ecology, pp. Excretion of orally administered -zinc-65 by wild 619-31in:Radioccology, V. Schultz and A.W. small mammals. Health Phys. H: 719-22. Klement, Jr. (Eds.). Reinhold. New York. Hagerman, F. B. 1952. The biology of the Dover sole Westrheim, S. J., and A.R.Morgan. 1963. Results from dlicrostoinus paciJicus. Calif. Fish and Game Fish. tagging aspawning stock of Dover soleAficrosiomrts Bull. 85: 1-48. paciftcus. Pac. Mar. Fisheries, Comm. Bull. 6: 14-21. Hoss,D. E. 1964. Accumulation of zinc-65 by flounder Winberg. G. G. 1956. [Rate of metabolism and food of the genus Paralichthys. Trans. Amer. Fish. Soc. requirementsoffishes.]Belorusskovo Nauchnye 93(4): 364-68. Trudy Belorusskovo Gosudarstvennovo Universiteta Mann,K. H. 1967. Cropping the food supply, pp. imcni V. 1. Lenina. Minsk. 235 pp. [Transl. Fish. Res. 243-57 in: The Biological Basis of Freshwater Fish Bd. Canada, No. 194.1 Production, S. D. Gerking (Ed.). Blackwell, Oxford. 126

RLO-2 22 7-T12-9

Losses of 65Zn to Inorganic Surfaces ina Marine Algal Nutrient Medium

Richard D.Tomlinson' and William C. Renfro Department of Oceanography, Oregon State University, Corvallis,Ore. 97331

(Na2Er rA) are often added to help maintain these nutrients The nature and magnitude of 85Z.n losses froma marine insolution, various workers (Chipman et al., 1958; Poli- algal nutrientmedium through adsorption to inorganic karpov, 1966; Bernhard and Zattera, 1967) have indicated surfaceswere examined. In the pH range 6.3-7.5, a precipitate that enrA renders zinc less available for algal uptake. There- formed in the medium. These particles accumulatedup to fore, any serious attempt to duplicate environmental condi- 70% of the "Zn in the medium within 24 hrat pH values tions for the study of zinc uptake rates in the laboratory must near 7.5. "Zn uptake at a pH of 6.3 was negligible. The use environmental zinc concentrations and avoid the addition filtered seawater base of the nutrient mediumwas used to of chelating agents to the artificial nutrient medium. Un- study "Zn losses to glassware surfaces. The relationship fortunately, omission of the chelator may engender chemical between the percent adsorption of "Zn froma contained precipitation, resulting in an increase of inorganic surfaces seawatersample and wetted-glass surface area/pipette sample available for adsorption of metallic cations such as zinc. volume was found to be linear for borosilicate glassvolu- There has been some concern that laboratory glassware metric pipettesin the size range tested (1-15 ml). At pH 8.0, might adsorb significant amounts of ionic zinc. Adsorption glasswarewith surface area/sample volume ratios as smallas of zinc by glass surfaces has generally been found negligible those of 20-m1 volumetric pipettes adsorbed 7--111 ofthe by workers using 15Zn as a tracer in relatively large containers containedsample activity. Use of polypropylene apparatus such as 125-nil borosilicate glass bottles (Schutz and Turekian, was found to significantly reduce zinc losses. It wascon- 1965) and 250-m1 borosilicate glass bottles (Robertson, 1968). cluded that "Zn adsorption by inorganic surfacescould However, it has not previously been determined whether result inserious errorsin measurementsof "Zn uptake by adsorption is significant when aqueous media having oceanic marine phytoplankton. zinc concentrations are transferred to containers with high surfacearea/volume ratios, such as pipettes. The work presented here represents an attempt to evaluate both quantitatively and qualitatively the loss of zinc from a L aboratory studies of "Zn uptake by marine algae have marine algal nutrient mediumto inorganic surfaces. generally shown large zincconcentration factors, where concentrationfactor is counts min-' gram-' fresh tissue/ Methods and Materials countsmin-' ml-' medium. Gutknecht (1965) observedcon- Algal Nutrient Medium.The algal nutrient medium used in centrationfactors of 30-3300 for "Zn accumulated by nine these studies was similar to that first characterized by Davis speciesof seaweed. On a tissue volume basis, Chipman et al. and Ukeles (1961). The onlyalterationsmade in the original (1958) found a'SZn concentration factor of 5 X 10' in cultures nutrientconcentrations were the elimination of Na2FDTA of the marine diatom,Nitzschia closteriurn.Through the use and ZnSO4, and the reduction of the FeCl3 concentration of carrier-free "Zn, they obtained evidencethatcell surface from 6.88 mg/l. to 1.28 mg/1. The medium base consisted of adsorption accounted for most of the zinc uptake. seawaterfiltered through a 0.45-µ pore membrane filter. In Adsorption of zinc from algal culture media isnot re- all experiments, sufficient carrier-free "Zn was added to the stricted to tissue surfaces alone. The formation of precipitates medium to givean initialactivity of approximately 4 µCi in artificialnutrient media may also contribute to the ad- "Zn/I. Table I lists the final concentrations of the various sorptiveuptake of zinc from solution. Nutrient media fre- chemical components added to the seawater base, for which quently contain abnormally high levels of nutrients incom- the concentration of total zinc was determined by atomic parison tothe natural environment (Fogg, 1966). Although absorption spectrometry to be 12.4 t 0.6 µg/l. at one stan- chelating agents such as sodium ethylenediaminetetraacetate dard deviation. Qualitative and QuantitativeAnalysis ofAlgal Nutrient MediumPrecipitate. When the newly made algal nutrient ' To whom correspondence should be addressed. medium was exposed to air, an increase in pH from 6.3 f 0.1

Volume 6, Number 12, November 1972 1001 127

TableI. Chemical Composition of Algal Nutrient Medium Table It.Qualitative and Quantitative Analysis of Algal Nutrient Medium Precipitate Component Concn° Physical characteristics of the precipitateas a function ofpH KH2PO4 200 mg/I. Relative NaNOs 150 mg/l. packed vol, NH4CI 25 mg/l. Al ppt/50 FeCls 1.28 mg/l. pH mlsuspension Color onsistency MnC12.4H20 360 µg/I. 6.3 f 0.1 1.5 red-brown gelatinous Thiamine HC1 200 eg/l. 6.7,± 0.1 2,0 red-brown gelatinous CoCI2.6H2O 22.0yg/1. 7.1 t 0.1 4.0 orange-browngelatinous to CuCI2.5H2O 19.5ug/l. particulate (NH4)2MoO4 12.5µg/1, 7.5 f 0.1 4.0 yellow-brownparticulate VitaminB12 2.0µg/I. Chemical compositionof precipitate formedat pH= 7.5 f 0.1 Biotin 1.0µg/1. Wt totalprecipitate Method of analysis ° Medium base: seawaterfiltered througha0.45-p poremembrane Component of filter. Pot 54.3 f 3.3° spectrophotometrye Ca 14.4 t 1.4 AASC Mg 6.8 t 0.3 AAS Fe 1.7 t 0.1 AAS to 7.5 f0.1 was observed. This change in pH was accompa- Zn 0.2 t 0.01 AAS niedby the formation of a flocculentprecipitate. Because the Mn not detectable AAS extent of 65Zn adsorption was found to bedependent on the Unknown 22.6 t 5.1 pH of themedium,the effect of thepH on the nature of the ° All error values represent estimates at I std dev. A Stricklandand Parsons(1965). precipitate was also examined. For this purpose, four 2-1. , Atomic absorption spectrometry. borosilicate glass beakers were filled withfreshly made non- radioactivenutrient mediumand sealed tightly with plastic lids and rubber gaskets to permitthe adjustment of the pH to four values held constant throughoutthe experiment. The well crystal coupled to a Nuclear Data Series 130 multi-channel initial pHvalues, attainedby bubbling air or pure CO2 analyzer. through the medium, were 6.3 t 0.1, 6.7 ± 0.1, 7.1 f 0.1, ADSORPTIONBY LABORATORYGLASSWARE. The stock solu- and 7.5 t 0.1. Homogeneity of the suspendedsolids was tion used inthe glasswareuptake studies was seawater maintained by continuous stirring. Samples were withdrawn filteredthrougha 0.45-µ poremembrane filter and spiked through small, otherwise-stoppered ports in the vessel lids. with carrier-free "Zn (4 eCi 65Zn/l., 12 µgtotal Zn/1.).Simi- The relative volume of precipitate in the samplecontainers larly preparedseawaterwas the base of thealgal nutrient was determined by centrifuging 50-m1 samplesand trans- medium used in the experiments previously described. Ad- ferring the solid portions to capillary tubes which had been sorptive uptake of 65Zn by borosilicate glassvolumetric sealed at one end and calibrated with mercury at successive pipettes was measured indirectlyby determination of the 65Zn/gram from the 1-µl intervals.The capillary tubes were then centrifuged decreasein thevalue of nCi solution, again, at 1500 rpm for 10 min, and their relative packed stock solution to the pipetted sample. The value ofnCi46Zn/ volumes were recorded. The precipitate formed at pH 7.5 t gram for the stock solutionwas established as an average 0.1 was analyzed for chemical compositionusing a variety for seven decay-corrected standards taken over a period of of methods which are listed with the accompanyingresults several days. Losses of 65Zn through adsorption to the walls in Table 11. Prior to analysis, the precipitate was collected of the stock solution container were foundto be negligible. by filtration, washed with deionized distilled water, and dried The standards were individually dipped outof the stock con- poured into a pre- at 110°C. tainer usinga polypropylene tube, and Zinc-65 UptakeStudies. ADSORPTION BY NUTRIENT MEDIUM weighed second tube. This sample was then weighed and PRECIPITATE. The uptake of 65Zn by thenutrient medium radioanalyzed, the volume having been adjustedto standardize precipitate was studied using four 1.5-1. suspensions having the counting geometry. The samples were pipettedinto pre- weighed, and radio- initialpH values of 6.3 f 0.1, 6.7 t 0.1, 7.1 f 0.1, and 7.5 t weighed polypropylene counting tubes, 0.1. The medium was spiked with 65Zn and keptin scaled analyzed. In each case, the sampleliquid was not allowed to beakers. rise above the volume mark on the pipette, so that only the For each sampling period, the radioactivity associated with surface area associated with the sample volumewas involved solution the precipitate was calculated as the differencein 65Znactivity in the adsorptive uptake process. The pH of the stock measured for equal volumes of the homogeneous medium was 8.0 ± 0.1. It was stirred throughout the experiment.For and its centrifugate. To minimize loss of 65Zn through ad- comparison of adsorptive potential, the geometricinner sur- sorption to the equipment, polypropylene centrifuge tubes face area of each pipette was approximatedas a combina- and radioanalysis counting tubes were used. Pipetting was done tion of cylinders and cones. with a 500-Al Eppendorf push-button microliter pipette with Results and Discussion disposablepolypropylenetips(Brinkmann Instruments, Westbury, N.Y.). Adsorption of 65Zn to the polypropylene Qualitative and Quantitative Analysis of Algal Nutrient surfaces in each case was found by direct measurement to be Medium Precipitate. The results of the qualitative and quanti- precipitate less thanI % of the contained sample. The accuracy of the tative examinations of thealgal nutrientmedium pipette was given by the mnnulncturcr as =1.0.3% at three are summarized in Table II. The ferric ion wasapparently standard dcviolions. responsible for giving the precipitate its red-brown to yellow- All samples were radioanalyzed in a 13,JC 13-cm Nal(T'I) brown color, since the panicles were found to be white when

1002Environmental Science & 't'echnology 128

The plots of pH vs. time in FeCl2 was not added. The color and consistencysuggest that are initial pH values in each case. an iron compound, perhaps hydrated ferric oxide, predomi- each figure give pH changes in the closedcontainers owing to The total natedat fill values of 6.3-6.7. addition of the 65Zn in acidic solution at time zero. The precipitate formed at pH = 7.5 consisted primarily of suspension sample for 128 hr in Figure ld was lost.The error orthophosphates. On the basis of its composition, low solu- brackets indicate a range of t 3 stddcv around the mean, bility, amorphous form (determined by X-ray diffraction) weighted according to the inverse of the individualsample and color, calcium orthophosphate [Ca;1(PO4)tj is a possible variances (Harley, 1967). suggests that the major constituent. High levels of phosphate are not unique to The decrease in pH seen in Figure Ic, d fraction 65Zn the modified Davis medium used in the present studies, but associated measurements of the particulate results seem to be relatively common in many nutrient media used for activity are conservative in comparison to projected algal culture. Accordingly, the utility of these observations for 65Zn uptake at a constant pH equal to theinitial value. 0.1 shows is not strictly limited to the single medium used in this study. Even so, the culture having an initial pH of 7.5 t Figure 1 completes the picture of the role of the nutrient an equilibrium adsorption levelof about 70% of the added medium precipitate in fiIZn adsorption. Figure la, b indicates 65Zn (Figure ld). The pH of this medium is near that of the Ii is clear that such a precipitate thatprecipitation in the pH range 6.3-6.7 results in very little growing culture of 1.galbana. depletion of soluble zinc levels. However, in an open con- can compete witha marine algain the accumulation of zinc tainer, the pH of the nutrient medium rose to nearly 7.5 in a from an aqueous medium. relatively short time. 'Table 11 shows that the relative packed Using similar levels of zinc-i.e., 10/J.g/l., Boroughs et at. volume of the precipitate increased by a factor of 2.7 between (1957) found that the marine diatom,N. closteriurn,removed pH values of 6.3 and 7.5. Simultaneously, the chemical nature nearly 90% of the zinc from a culture in 4 hr. At pH 7.5 -L of the precipitate changed and its capacity to adsorb 65Zn 0.1, the present studies show that a nutrient mediumprecipi- zinc in the same increased.In anopen culture of theyellow-green alga, Iso- tate is capable of taking up over 50% of the chrysis galbana,pH values remained withinthe range7.4-7.6 lengthof time. This comparison, though somewhat limited owing to lack of pH and particle density data for thealga, from inoculation to senescence. in Figure 1 shows an increase in adsorption of 65Zn by the suggeststhat the precipitate and the alga could be present precipitate, with increases in pH over the range 6.3-7.5. These a single culture, each accumulating zinc.

z (a) ^o0 role/ Susprnsion 4 ------rote/ Suspension I00 W W Y-- CenJriJu olr N !J

Clneri/ug4se

3000 E3000 a 1000

PertCU/att Fraction Port/cubte FrocNdn n 6

640 a. 680 n G 6.60 6.20 40 60 80 100 120 0 20 40 60 80 100 120 0 ELAPSED TIME (hours) ELAPSED TIME (hours) (d)

Total` Suspension .100 4000 7 0 2 1-60 14 Particulate Fraction in. N (D) W a E 2000

N Centrifugal* e 1000

7.40

M7.20

..w 60 60 90 100 20 40 60 60 100 120 00 20 4040 ELAPSED TIME (hours) ELAPSED TIME (hours) Figure 1. Effect of pi I on 61Zn uptake by a precipitate in a marine algal nutrient medium Initial pit values of the suspensions are (a) 6.3 a_ 0.1, (b) 6.7 f 0.1,(c) 7.1 t 0.1. and (d) 7.5 t 0.1. Excepting Ofvalues, each data point represents the mean result of two individual anal) ses

Volume 6, Number 12, November 1972 1003 1.'9

12 osm1 °smi e3 The effect of pretreatment of the pipetteswith dilute HF may r------zomi° 2,1 be seen in a comparison of the results of this experiment (Fig- z 4 4 (a) v.001x.9.97 ure 2b) with those previously examined (Figure 2a). HF ire- .09 o treatment of previously used pipettes helped toeliminate 0 individual pipette surface variations (arising from varied (b) adsorption histories) and permitted collection of adsorption data on a standard basis. With respect to adsorptive uptake of zinc by borosilicate V ..013 x ..300 glass volumetric pipettes, several pertinentobservations cat. 93 be made on the basis of data presented in Figure 2. For the 0 400 8To 1250 relationship examined, percent sample adsorptionvs. glass SURFACE AREA / VOLUME (mm2 / ml ) surface area/sample volume, HF pretreatment increased the Figure 2. Percent adsorption of Zn from seawater samples by correlation ofX- Yvalues (in comparison to acetone pretreat borosilicate glass volumetric pipettes as a function of pipette surface ment) and placed the Y-intercept nearer to the theoretical area/'contained sample volume zero value. The relationship is confirmedas linear for the (a) HCI and acetone pretreatment of glass surfaces, (b) HF pretreatment of range of values tested. glass surfaces Because sample volume is directlyproportionalto total available zinc, zinc adsorption by borosilicateglass is therefore shown to be a function of the availablezinc and the glass sur- Under such conditions considerable error could result from face area contacted by the liquid. the assumptionthatloss of zinc from the culturemedium was For the conditions used(12 µg totalzinc/I., pH - 8.0 f due to uptake by the alga alone. This errormight be elimi- 0.1), zinc adsorption by previously used pipettesof the size nated through the use of controlsrequiringcareful pHregula- range, 2-20 ml, amounts to 7-11 % of the zincin the contained tion. Because the precipitationin the culturesis a function of their pH, the pH of the control medium would have to be sample (Figure 2a). From a practical viewpoint, these results showthat zinc continually adjusted to match thatof thecultures. This might adsorption onto glass from a seawatermedium is negligible prove difficult in experiments where pHchangesare more de- for glass laboratory apparatus havinga low surface area/ pendent on photosynthesisand respiration than onmechan- volume ratio. This includes large beakers, flasks,graduated ical aerationand mixing. cylinders, etc. This observation agrees with the results ob- Although the use of controlsin such experiments would tained by Robertson (1968) for 200-mlborosilicate glass permit a quantitative uptake evaluation,zinc accumulation bottles. However, adsorption of zinc bysmall glassware such rates and equilibrium levels measured under such conditions as pipettes may result in serious error. At pH 7.5, which was would likely differ significantly from those measured in cul- the average value measured for the growing algal culture of 1. tures free from inorganic particulate matter. It is well docu- zinc adsorption by the 10-ml pipettein Figure 2a mented that zinc uptake by marine algae is affected by the galbana, would amount to only about 2 %. This can be determined by levels of zinc in solution (Knauss and Porter, 1954; Townsley examination of the linear pH-adsorptionrelationship out- et at., 1959-1960). lined by Tomlinson (1970). For smallerpipettes this error Potential zinc uptake by inorganic solids may also be found would be potentially larger, and would increasefor all boro- significant in heat-sterilized seawater media, regardless of silicate glassware with an increase in pH. At natural oceanic added nutrient levels. The findings of Jones (1967),showing pH values of about 8.0, zinc adsorption errors may be of the up to 20 mg/I. of precipitate in autoclaved natural seawater order of 7-11 % for even relatively large pipettes.Such errors (pH = 7.9), suggest this possibility. can be reduced considerably by a minimum of two prerinses "Zinc Uptake Studies: Adsorption by Laboratory Glassware. with the sample liquid and/or treatment of the glasssurfaces A linear relationship between the two parameters, percent sample adsorption and exposed glass surface area/pipette with various commercial chemicals which produce hydro- phobic effects (Tomlinson, 1970). It should be observed that volume, was obtained by the method of least squares, with no attempt to force the line through zero (Figure 2a). The low the latter method may necessitate recalibration of volumetric glassware owing to alteration of the shape of theliquid me- correlation of values and the failure of this plot to intercept the theoretical value of zero was attributed to the varied niscus and the reduction of surfacewetting. adsorptive history of the pipettes. To givethis experiment a Acknowledgment practical aspect, these pipettes were chosenat random from a stock of previously used laboratory equipment. They were The authors thank K. L. Carder, N. H. Cutshall, W. O. prepared by rinsing with 6N HCI, distilled deionized water, Forster, and L. F. Small for the time and effort that they and acetone. Figure 2a indicates that this treatment,a common devoted to this work. glassware cleaning technique, was inadequate to remove previously adsorbed substances from the glass surfaces. Literature Cited The relationship depicted in Figure 2a was reexamined by Bernhard, M., Zattera, A., "Proceedings of the Second rinsing a similar set of pipettes with dilute HF priorto expo- National Symposium on Radioecology," Ann Arbor, Mich. (usACC CONE-670503), 1967. sure to the radioactive solution. Tomlinson (1970) showed Boroughs, H., Chipman, W. A., Rice, T. R., "The Effects of that pretreatment of borosilicate glass surfaces with 0.3N HF Atomic Radiation on Oceanography and Fisheries," NAS- had no significant effect on the time required to establish NRS, Washington, D.C., .NAS-NRC Publ. No. 551, pp 80--7, equilibrium of zinc uptake from solution. Quantitatively, how- 1957. ever, the equilibrium level of adsorbed zinc was increased by a Chipman, W. A., Rice, T. R., Price, T. J., U.S. Fish and Wildlife Service, Fishery Bull. 135, p 279-92 (1958). factor of five under these conditions. The glass surfaces used Davis, H. C., Ukeles, It., Science, 134, 562-4 (1961). in his experiments had not previously been exposed to an Fogg, G. E., "Algal Cultures and Phytoplankton Ecology," aqueous medium. Univ. of Wisconsin Press, Madison. Wis.. 126 pp, 1966.

1004Lnvironmenlal Science & Technology 130

Gutknecht, J., Limnol. 'Oceanogr., 10, 58-66 (1965). Tomlinson, R. D., MS thesis, Oregon State University, Harley, J. H., Ed., "Manual of Standard Procedures," 2nd Corvallis, Ore.;'1970. ed., New York, N.Y., USAEC (NYO-4700), 1967. Townsley, S. J., Reid, D. F., Ego, W. T., "Uptake of Radio- Jones, G. E., Limnol. Oceanogr., 12, 165-7 (1967). isotopes and Their Transfer Through Food Chains by Knauss,H. J., Porter,J. W., Plant Physiol., 29, 229-34 (1954). Marine Organisms," 40 pp, Anna. rept. (1959-60), Univ. Polikarpov, G. G., "Radioecology of Aquatic Organisms," of Hawaii contract At(04-3)-56 with U.S. Atomic Energy North-Holland, Amsterdam, 314 pp. 1966. Comm. (microfilm). Robertson, D. E., Anal. Chin. Arta, 42, 533-6 (1968). Schutz, D. F., Turekian, K. K.,Geochim Cosmochim. Aeta, Received for review September 13, 1971. Accepted June 26, 29, 259 (1965). 1972. This work =was supported by funds made available under Strickland, J. D. H., Parsons, T. R., "A Manual of Sea Water Atomic Energy Commission Contract no. At(45-1)2227, Task Analysis," 2nd ed., Ottawa, 203 pp (Fish. Res. 13d. Can., Agreement 12 (RLO-2227-T12-9) and Federal Water Pollution Bull. No. 126), 1965. Control Administration Grant No. wp 111-04. 1.31

RLO-2227-T12-13

OPHEL..IA, 10: 35-47 (August 1972)

FOOD SOURCES OF SUBLITTORAL, BATHYAL AND ABYSSAL ASTEROIDS IN THE NORTHEAST PACIFIC OCEAN

ANDREW G. CAREY, JR. Department of Oceanography Oregon State University, Corvallis, Oregon 97331, U. S.A.

ABSTRACT Sea stars from the sublittoral to the abyssal zone at 4260 meters depth in the northeast Pacific Ocean were examined for food source. A total of 491 specimens of 29 species were dissected for stomach content analyses. Seven species are predators; six, deposit-detritus feeders; thirteen, omnivores; and the feeding habits of three remain unknown. There are changes in feeding type with increasing depth; the relative abundance of predators decreases from 67 to 0 %, and the deposit-detritus feeders, from 33 to 14 %. The omnivores increase from 0 % on the inner continental shelf at a depth of 50 meters to 71 % on the abyssal plains. It is concluded that in the near-shore abyssal environment the deep-sea asteroids are generally facultative feeders and obtain food from the sediments as well as prey or animal remains, while in the food-rich shallow waters of the inner continental shelf the asteroid fauna are generally specialized carnivores.

INTRODUCTION One of the major interactions between organisms in the sea is the search for, and ingestion of, food. The utilization of energy and the role in the cycling of elements by benthic animals in the marine ecosystem can be understood only when food sources of each dominant species are known. As an initial step in linking the Oregon benthic fauna together in a functional manner, feeding patterns of sea stars have been studied. Though observations on the feeding of asteroids have been made for a long time (see Hyman, 1955 and Feder & Christensen, 1966, for reviews), the food sources of many species, particularly deep-sea forms, have remained unclear. It is necessary to observe the animals feeding directly (Mauzey, Birkeland & Day- ton, 1968) or to obtain a large series of stomach content analyses in order to unravel this segment of the benthic food web. The major sources of food for certain sublittoral, bathyal, and abyssal asteroids off the coast of Oregon are described by analyzing a large series of asteroid stomachs. The range of depths and types of environments covered is broad and should allow one to make gene- ralii.ations concerning changes in food source with depth.

3* 132

36 ANI)RI?W G. CAREY, JR. The species composition and depth distribution of the asteroid fauna from sublittoral and bathyal zones off northern Oregon have been reported and summarized by Alton (1966). McCauley (1972) summarized the species collected during Oregon State University oceanographic cruises and listed their depth distributions. The feeding habits and behavior of shallow subtidal asteroids in the Pacific Northwest have been well characterized by Mauzey, Birkeland & Day- ton (1968) from a large series of direct observations by SCUBA divers. The present paper increases the depth range covered and describes the major trends in food sources with increasing depth and distance from shore.

The field assistance of D. R. Hancock and R. R. Paul and the laboratory assistance of Mrs. G. S. Alspach and G. L. Hufford are gratefully acknowledged. M. S. Alton, M. Downey, F.J. Mad- sen, and J. E. McCauley kindly undertook some of the identifications. Helpful comments on the manuscript were made by J. E. McCauley, R. E. Ruff, and F.J. Madsen. Several asteroid collections were provided by W. G. Pearcy and J. E. McCauley. The financial assistance of the Atomic Energy Commission (Contract AT(45-1)2227-Task Agree- ment No. 12) and the National Science Foundation (Grants GB-4629 and GB--53I) are gratefully acknowledged. The National Science Foundation (Grant GD-27257) provided partial support for ship operations. The manuscript (RLO-2227-T12-13) was completed at the Zoo- logical Museumof the University of Copenhagen in 1970 under partial support of a George C. Marshall Memorial Fellowship in Denmark.

MATERIALS AND METHODS The specimens came from otter trawl samples obtained off central and northern Oregon (Carey, 1968, 1972). Seven stations off central Oregon from 50 to 2860 meters depth were sampled seasonally from1962to 1968. Other stations further from shore, at intermediate depths, and off northern Oregon were occupied from time to time. A 7-meter semi-balloon Gulf of Mexico shrimp trawl was the standard sampling gear. The 3.8 cm stretch mesh usually was fished with a 1.3 cm stretch mesh liner in the cod-end. Samples were sorted into major taxonomic categories on the ship, and asteroids to be used for stomach content analyses were separated out and preserved in 10 % formalin neutralized with sodium borate. Concentrated neutralized formalin was injected into the stomach when possible. Some specimens, deep-frozen at sea for radioecological research, were later thawed and dissected in the laboratory. We have collected 70 or more species of Asteroidea in the northeastern Pacific Ocean (McCauley, 1972); 29 species were analyzed for stomach contents. Only the more abundant species collected during the course of this study and ones not needed for other research projects could be investigated for food source. The contents were carefully washed from the stomach cavity through a circular incision on the dorsal side of the starfish. The contents wered store separately i 70 % isopropyl alcohol in glass vials and were then examined under a dissecting microscope and enumerated for each specimen. A rough volume scale was used 1 3 3

FOOD SOURCES OF ASTEROIDS 37 to note the volume of the food item in the stomach: trace: very small amount; some: full; much: > J to a full; and maximum: a to full. The starfish were weighed and measured. Stomach contents were summarized for each species, and conclusions drawn about the major source of food. Data from pertinent literature have been included.

RESULTS The 29 species (491 specimens) of asteroids cover a broad depth range from 46 to 4260 meters (Fig. 1). The number of species is greatest on the upper continental slope. In general there is a continuous change in the species composition with depth, and the upper slope is a transition zone in the asteroid distribution. A number of bathyal species reach up to 200-400 meters depth, while much of the shallow water fauna is limited to the continental shelf or just beyond. Stomach contents and feeding types are summarized by species; the sea stars are classified to feeding type by synthesizing the present results with those obtained from the literature. Species classified as predators, mostly sublittoral forms, almost always contain plentiful small prey or fragments of larger animals. The species classified as deposit-detritus feeders consistently contained sediment; some had only a small amount while others were packed full. Meiofauna and animal fragments were mixed with the sediment in about the same proportion as in the surrounding surface sediments. When a small amount of fine sediment was consistently present in the stomach, particularly if it was coagulated with a mucous-like material, it was inferred that the organisms had been feeding on detrital organics on the sediment surface. Some species appear to feed both upon sedimentary organics and on other organisms. They have been classified as omnivores because the stomachs generally contained sediment with a considerable number of animal fragments. They may be scavengers rather than active predators, but their diet is a varied one in any case. They appear to feed on whatever organic material is available. Although I have classifiedZoroaster evermanniFisher andZ. ophiurusFisher as omnivores, Alcock (1893) noted that another species from the deep-sea,Z. carinatusAlcock, feeds largely on molluscs and crustaceans. Data from the present studies indicate thatSolaster borealisFisher is probably an omnivore for it generally contains sediment as well as animal fragments. However, four other species ofSo- lasterhave been classified as carnivores by other workers:S.dawsoniVerrill eats mainly asteroids whereasS.stimpsoniVerrill eats mainly holothurians (Mauzey et al., 1968).S.endeca(Linnaeus) feeds on sea anemones, asteroids, and holothurians (Grieg, 1913; Blegvad, 1914; Bull, 1934; and Mauzey et al., 1968); and in one caseS.paxillatusSladen had ingested a holothurian(Molpadia intermedia Fisher, 1911). In the northeastern Pacific OceanHymenaster quadrispinosusFi- sher, an omnivore, feeds on sedimentary organics as well as on some organisms; 134

38 ANDREW G. CAREY, JR.

DEPTH IN METERS

SPECIES AO'_ 2 4000 T RkOO 2851 1 Mediaster oequalL Pisaster brevisp/nus Luidia folio/ata Pseudorchaster pare/i/ a/ascensis Sty/osterias forreri Hippasteria spinosa Dip/opteraster mu/t/pes Rothbunoster ca/ifornicus Ctenodiscus cr/spotus So/aster borealis Pteras/eridoe sp Nearchoster ocicu/osis Thrissicanth/as pencil/atus Heterozonias olternatus Dipsacoster anop/us Zoroaster evermanni Hippaster/os ca/iforn/cus Pedice/lasterinoe sp Zoroaster ophiurus Lophoster furcIlliger Hymenoster quadrispinosus Hymenoster sp P1/aster pec/inatus Pectinaster sp Pseudarchaster dissonus Dytaster sp Mediaster e/egans abyssi Benthopecten sp Eremicaster pocificus

DEPTH DISTRIBUTION KNOWN DEPTH DISTRIBUTION OF DISSECTED ASTEROIDS 0OFF OREGON (McCouIey, in Press) Ftc. 1. Depth distribution of asteroid species off Oregon studied for food source. The depth range of the dissected specimens and the known range off Oregon coast (McCauley, 1972) are indicated.

whereas H. blegvadi Madsen eats small snails and possibly clams (Madsen, 1956) and H.pellucidus Wyv. Thomson ate an ophiuroid in one case (Mortensen, 1927). Mauzey et al. (1968) report that Hippasteria spinosa Verrill is, in shallow water, clearly a predator on the pennatulid anthozoan Ptilosarcus gurneyi. Ob- servations by SCUBA divers demonstrated that Ptilosarcus retracts into the sediment when attacked by Hippasteria. The asteroid then digs out the anthozoan and digests the apical end of the prey with its extruded stomach. These data strongly suggest that Hippasteria spinosa have similar feeding habits on penna- tulids on the central Oregon upper continental shelf. 1:35

FOOD SOURCES OF ASTEROIDS 39 The stomachs of four species either contained so little material or the sample number of starfish was so small that determination to feeding type was impossible. It is possible that they digest prey extra-orally with an everted stomach, a type of feeding that is very difficult to detect by stomach content analyses. Eight specimens out of nine ofPectinastersp. examined for stomach contents were empty; the remaining one contained but a trace of sediment. Alcock (1893) has reported thatP. hispidusAlcock & Wood-Mason, another deep-sea species, feeds mainly on molluscs and crustaceans. It may be that the speciesof Pecti- nasterstudied here is a predator digesting prey extra-orally or feeding infrequently, but the data are too few to permit conclusions. The dominant food sources for each species are given in italics in the following list in order of their depth distribution. The occurrence of one of a food item is indicated by the singular, and more than one by the plural. Mediaster aequalisStimpson. Depth range of specimens: 46 to 200 meters. Number dissected, 9; number with material in stomach, 5. Stomach contents: fecal pellets and trace to muchsediment.[Literature: detritus or decaying animals (Hopkins & Crozier, 1966), sponges and ectoprocts on rocky substrate; sea pens and drift algae on sandy substrate; sediment on muddy substrate (Mauzey et al., 1968)]. Feeding type off Oregon: deposit-detritus. Pisaster brevispinus(Stimpson). Depth range of specimens: 50 meters. Num- ber dissected, 4; number with material in stomach, 2. Stomach contents: some sediment. [Literature: pelecypods, on sandy substrate, barnacles on rocky substrate (Feder & Christensen, 1966; Mauzey et al., 1968). Feeding type off Ore- gon : predator. Luidia foliolataGrube. Depth range of specimens: 56 to 245 meters. Number dissected, 26; number with material in stomach, 24. Stomach contents: pelecypod, echinoids, holothurian,ophiuroids,polychaete, crustacean, and some sediment. [Literature: pelecypods, holothurians, crustaceans (Mauzey et al., 1968); ophiuroids and schaphopods(Dentalium)(Fisher, 1911)]. Feeding type off Ore- gon : predator. Pseudarchaster parelii alascensisFisher. Depth range of specimens: 165 to 400 meters. Number dissected, 12; number with material in stomach, 9. Stomach contents:gastropods, ophiuroids,crustaceans, foraminiferan, and trace to maximumsediment.[Literature: none]. Feeding type off Oregon: omnivore. Stylasterias forreri(de Loriol). Depth range of specimens: 200 meters. Num- ber dissected, 1; number with material in stomach, 1. Stomach contents:gastropods (5)and some sediment. [Literature: gastropods, chitons, diatoms, and ectoprocts (Mauzey et al., 1968)]. Feeding type off Oregon: predator. Hippasteria spinosaVerrill. Depth range of specimens: 200 to 800 meters. Number dissected, 3; number with material in stomach, 3. Stomach contents: trace to some sediment, [Literature:pennatulidsand occasionally sea anemones and gastropod eggs (Mauzey et al., 1968)]. Feeding type off Oregon: predator. 136

40 ANDREW G. CAREY, JR.

Diplopteraster multipes(Sars). Depth range of specimens: 400 meters. Number dissected, 2; number with material in stomach, 2. Stomach contents: ophiuroid- like fragments and some sediment (sand). [Literature: none]. Comments: the stomach was partially everted in both; this species could be a predator with extra-oral digestion. Feeding type off Oregon: unknown. Rathbunaster californicusFisher. Depth range of specimens : 400 meters. Num- ber dissected, 2; number with material in stomach, 1. Stomach contents:echinoid fragments,crustaceanfragments, and sediment. [Literature: crustaceans in the laboratory (Fisher, 1928)]. Feeding type off Oregon: predator. Ctenodiscus crispatus(Retzius). Depth range of specimens : 400 to 1600 meters. Number dissected, 27; number with material in stomach, 26. Stomach contents: a pelecypod, a foraminiferan, spicules, worm tubes, and some to maximumsediment.[Literature: mud (detritus) (Fisher, 1911; Gisln, 1924; and Mortensen, 1927)]. Comments: mostspecimenshad much to maximum sediment in stomach. Feeding type off Oregon: deposit-detritus. Solaster borealisFisher. Depth range of specimens: 400 to 1600 meters. Number dissected, 23; number with material in stomach, 14. Stomach contents: echinoidfragments,crustacean, trace to somesediment.[Literature: none]. Com- ments :most specimens had sediment in stomach. Mouth wide open in two, and unidentified material in three. Feeding type off Oregon; omnivore. PTERASTERIDAE. Depth range of specimens: 400 to 1600 meters. Number dissected, 6; number with material in stomach, 6. Stomach contents:ophiuroidand echinoidfragments, fecal pellets, and trace to maximumsediment.[Literature: none]. Feeding type off Oregon: omnivore. Nearchaster aciculosus(Fisher). Depth range of specimens: 400 to 2086 meters. Number dissected, 32; number with material in stomach, 25. Stomach contents: fish scale, crustaceans, ophiuroids, fecal pellets, and trace to muchsediment.[Literature: none]. Comments: mouth open in two, unidentified organic matter in four. Feeding type off Oregon: omnivore. Thrissacanthias penicillatus(Fisher). Depth range of specimens: 540 to 1200 meters. Number dissected, 31; number withmaterial instomach, 26. Stomach contents:pelecypods,echinoids, ophiuroids, crustaceans, scaphopod,gastropods, and trace to much sediment. [Literature: none]. Comments:gonads mature in two specimens from a June sample. Feeding type off Oregon: predator. Heterozonias alternatus(Fisher). Depth range of specimens: 600 to 1200 meters. Number dissected, 21; number withmaterial instomach, 15. Stomach contents: crustacean, foraminiferan, sea pen stalk (?), and trace to somesediment.[Literature: none]. Comments: 11 specimens containedsediment.Feeding type off Oregon: deposit-detritus. Dipsacaster anoplusFisher. Depth range of specimens: 600 to 2800 meters. Number dissected, 44; number with material in stomach, 41. Stomach contents: pelecypods, gastropods, ophiuroids,crustacean, polychaete, some to much sedi 13'

FOOD SOURCES OF ASTEROIDS 41 ment. [Literature: none]. Comments: specimens from 600 to 800 meters contained more animals. Feeding type off Oregon: omnivore. Zoroaster evermanniFisher. Depth range of specimens: 800 meters. Number dissected, 2; number with material in stomach, 2. Stomach contents: echinoderm fragment, spicules,crustacean(one large shrimp), isopod, andsediment.[Liter- ature: none]. Feeding type off Oregon: omnivore. Hippasteria californicaFisher. Depth range of specimens: 800 to 1260 meters. Number dissected, 9; number with material in stomach, 3. Stomach contents: trace to somesediment.[Literature: none]. Feeding type off Oregon: deposit- detritus. PEDICELLASTERIDAE. Depth range of specimens: 800 to 1260 meters. Number dissected, 18; number with material in stomach, 11. Stomach contents :fecal pellets and trace to somesediment.[Literature: none]. Comments: mouth open in five; sediment coagulated in some. Most animals had some sediment. Feeding type off Oregon: deposit-detritus. Zoroaster ophiurusFisher. Depth range of specimens: 800 to 2176 meters. Number dissected, 5; number with material in stomach, 3. Stomach contents: ophiuroidand crustacean fragments, and some to maximumsediment.[Literature: none]. Feeding type off Oregon: omnivore. Lophaster furcilligerFisher. Depth range of specimens : 1097 to 2176 meters. Number dissected, 29; number with material in stomach, 20. Stomach contents: echinoidandophiuroidfragments, worm tube, foraminiferan, and trace to maximumsediment.[Literature: none]. Comments: mouth open in three, stomach partially everted in four. Feeding type off Oregon: omnivore. Hymenaster quadrispinosusFisher. Depth range of specimens: 1540 to 2926 meters. Number dissected, 47; number with material in stomach, 27. Stomach contents: crustacean andechinoidfragments, gastropod, and trace to muchsediment.[Literature: none]. Comments: amorphous organic material in nine, mature gonads in one from a June sample. Feeding type off Oregon : omnivore. Hymenastersp. Depth range of specimens: 1600 meters. Number dissected, 2; number with material in stomach, 2. Stomach contents:echinoidandophiuroid fragments, some to muchsediment.[Literature: none]. Feeding type off Oregon: omnivore. Psilaster pectinatus(Fisher). Depth range of specimens: 1800 to 2500 meters. Number dissected, 16; number with material in stomach, 11. Stomach contents: some to maximumpelecypods, scaphopods,gastropod, polychaetes, echinoid, holothurian, and trace to much sediment. [Literature: pelecypods and scaphopods (Sokolova, 1957)]. Feeding type off Oregon : predator. Pectinastersp. Depth range of specimens: 2176 meters. Number dissected, 9; number with material in stomach, 1. Stomach contents: trace of sediment. [Literature: none]. Comments: amorphous material in one; mouth open in six. Feeding type oil' Oregon: unknown. 138

42 ANDREW G. CAREY, JR. Pseudarchaster dissonusFisher. Depth range of specimens: 2176 to 2850 meters. Number dissected, 18; number withmaterial instomach, 9. Stomach contents:ophiuroidfragments,echinodermandspongespicules, and trace to maximumsediment.[Literature: none]. Comments: mouth open in six, stomach everted in four. Feeding type off Oregon: omnivore. Dytastersp. Depth range of specimens: 2500 to 4260 meters. Number dissected, 24; number with material in stomach, 11. Stomach contents:echinoid andophiuroidfragments,spongespicules, and some to maximumsediment.[Liter- ature: none]. Comments: most with much sediment. Feeding type off Oregon: omnivore. Mediaster elegans abyssiLudwig. Depth range of specimens: 2600 to 3050 meters. Number dissected, 19; number with material in stomach, 5. Stomach contents: siliciousspongespicules, tube, and trace to some sediment. [Literature: none]. Comments: mature gonads in two (December and February samples). Feeding type off Oregon: unknown. Benthopectensp. Depth range of specimens: 2600 to 3700 meters. Number dissected, 44; number with material in stomach, 15. Stomach contents:ophiuroid fragments, isopod, and trace to somesediment.[Literature: none]. Comments: mouth open in three, stomach everted in four. Feeding type off Oregon: omnivore. Eremicaster pacfcus(Ludwig). Depth range of specimens: 2772 to 2860 meters. Number dissected, 6; number with material in stomach, 6. Stomach contents: foraminiferans, much to maximumsediment.[Literature: none]. Comments: some amorphous material in stomachs. Feeding type off Oregon: deposit-detritus. Depth trends are evident in the food sources of Oregon asteroids (Fig. 2). Predators, relatively more abundant on the continental shelf, become increasingly less prevalent with depth. No sure predators are present among the species studied from the abyssal environment. Deposit and detritus feeders comprise a fairly constant portion (24-40 %) of the asteroid fauna from theinnershelf to a depth of 1200 meters on the upper slope. Beyond this depth the asteroids living solely on food materials associated with the sediment decline to 14-18 ° of the sea star fauna. In contrast, omnivores are not present on theinnershelf, but comprise 20 % of the species on the outer shelf and continuously increaseto a maximum of 71 % on the abyssal plains. 1;39

FOOD SOURCES OF ASTEROIDS 43

ASTEROID FEEDING TYPES spp. INNER SHELF O-50m (3)

OUTER SHEL F >50-200m (5)

UPPER SLOPE >200-1200m (I71

LOWER SLOPE >1200-2600 (/7)

ABYSSAL PLAIN >2600-4260m (7) 0 20 40 60 80 100 FEEDING TYPES

nPREDATOR DEPOSIT AND DETRITUS OMNIVORE UNKNOWN FIG. 2. Summary of trends with depth of asteroid food source in the Northeast Pacific Ocean. When a species is present in more than one depth zone, it has been included in both areas.

DISCUSSION The species composition of the asteroid fauna off Oregon changes with depth (Fig. 1 and McCauley, 1972). There are gradients also of temperature, salinity, dissolved oxygen, pressure, and sediment particle size that change together with increasing depth. These gradients affect the distribution of single species, but generally do not act as a barrier to a large portion of the sea star fauna. There is, however, a large change in the species composition in the depth interval between the outer continental shelf and upper continental slope. The transition from the shallow, seasonally controlled sublittoral environment to the more stable, deeper environment of the bathyal and abyssal zones probably accounts for the marked change in species composition (Sanders & Hessler, 1969). Alton (1966) has reported a similar increase in the number of species of sea stars on the northern Oregon continental slope. Below the upper bathyal zone, asteroid spe- 140

44 ANDREW G. CAREY, JR. cies appear to react individually to the change in benthic environment with depth. The general trend noted in the diet of Oregon asteroids has been reported by Thorson (1957) for other benthic organisms; the incidence of predation decreases with depth. Sokolova (1959), reporting similar trends over very broad depth ranges (from sublittoral to hadal depths), related the trophic structure of the faunal assemblages to sediment type, detrital food supply in sediment and bottom water, and the character of near-bottom currents. When there is sufficient detrital and sedimentary organic material, the benthic fauna is composed of trophic groups that feed upon these materials. When there is insufficient food material associated with the sediments, selective detritus feeders are absent and filter feeders predominate. Based upon the predominance of deposit-detrital and omnivorous feeding asteroids and of deposit-feeding infauna (Carey, 1965), the nearshore Cascadia Abyssal Plain corresponds to the eutrophic abyssal structure described by Sokolova (1966). As food is probably limiting in the abyssal environment and as it probably reaches the sea floor as particulate material (Ekman, 1953; Sanders & Hessler, 1969; and Carey, unpublished), it is likely that organisms will become lessspe- cialized feeders and will tend toward deposit-detrital and omnivorous feeding. They will feed on whatever utilizable organic materials theycan get. Thus, food quality and supply is a regulator of faunal abundance and energy use in the deep- sea. The availability of types of food, prey or non-living organics, appears to influence the trophic structure at depth; when prey is not available, the omnivores eat what they can get. Feder & Christensen (1966) found that the majority of all sea stars studied up to that time were carnivores that only occasionally actedas scavengers. The majority of these asteroid species are from littoral or sublittoral environments. At bathyal and abyssal depths, however, predators are fewor absent; deposit- detritus feeders and omnivores predominate. In the present study, thirteen of twenty-nine species studied resort to ingestion of sediment in significant quantities as well as animals or their remains, and should be characterized as omnivores rather than predators. A number of starfish have been reported as omnivorous scavengers feeding by flagellary mucoid means (MacGinitie & MacGinitie, 1949; Anderson, 1959; and Pearse, 1965). The distinctions between carnivore, omnivore, and omnivorous scavenger are probably morea matter of degree and interpretation than clear-cut categories in manycases. The deposit-detritus category combines in the presentpaper at least two types of feeders, the ingesters of whole sediment and those which feed exclusivelyon the detritus found on the sediment surface. Sokolova (1958) separated thenon- selective detritus feeders (i. e. deposit feeders) from the selective detritus feeders that collect detritus from the sediment surface. In the present study, the two types could not be clearly distinguished. Concerning the Porcellanasteridae, to which 141

FOOD SOURCES OF ASTEROIDS 45

Eremicasterbelongs, Madsen (1961) states that besides being deposit feeders they may also be scavengers and facultative predators. Stomach content analyses present a number of problems for reconstructing the food web. Although deductions can be made concerning the major food sources in manycases,nothing can be determined about the rates of ingestion andassimilation.Deterioration of food items in the stomach from digestive processes during sample retrieval and before preservation of the specimen can cause problems.Luidia sarsiDuben & Koren digests its ophiuroid prey in 16 to 24 hours at 12-13'C (Fenchel, 1965), leaving ample time for collection and preservation. However,Christensen (1970) has demonstrated that rates of digestion in a shallow water carnivorousasteroid, Astropecten irregularis Pennant,are correlated with the palatability of the food species. Digestion can last for several hours to several weeks depending on the resistance of the prey to anaerobic conditions. How many species in the food-limited bathyal and abyssal environments exhibit such specific dietary preferences and food manipulations is not known, but such possible complex behavior must be considered. Sokolova (1957) reports evidence from stomach content and faunal analyses, that some deep-sea benthic organisms select their prey. During retrieval of the trawl sample from the seafloor,some species may undergo muscle spasms fromthermal, mechanical,or pressure shock and empty their stomachs (cf.Christensen,1970). Extrusion of the stomachs with extra-oral digestion can also createdifficulties,for it is most difficult to detect by stomach contentanalysis.Other types of behavior such as ciliary or flagellar mucoid feeding onsmall particles(Anderson,1960) yield only traces of material in the stomach. Direct observationin situand subsequent retrieval and laboratory experimentation can solve many of these problems (Mauzey,Birkeland, & Dayton,1968), but this is not yet feasible on a broad scale with the deep-seafauna.Bottom photography can show certain types of behavior that aid in interpreting the stomach content data, but a series of photographs from the Oregon sampling stations have failed to demonstrate prey-predator or other feeding relationships. In spite of these and other difficulties, stomach content analyses can yield valuable results that add to our knowledge of the interactions of the benthic fauna. Inasteroids,there is a shift from a specialized to a more generalized diet, with the food source changing from animal prey to sedimentary organics and animal remains.Asteroids,like many other abyssal fauna groups, in most cases ingest and assimilate whatever utilizable food materials are available in the food- limited deep sea. 142

46 ANDREW G. CAREY, JR.

REFERENCES ALCOCK, A. 1893. An account of the collection of deep-sea Asteroidea. Natural History notes from H. M. Indian Marine Survey Steamer INVESTIGATOR. Ann. Mag. nat. Hist., 6. Ser., 11: 73-121. ALTON, M. S. 1966. Bathymetric distribution of sea stars (Asteroidea) off the Northern Oregon coast. J. Fish. Res. Bd Can., 23: 1673-1714. ANDERSON, J. M. 1959. Studies on the cardiac stomach of a starfish,Patiria miniata(Brandt). Biol. Bull. mar.biol. Lab., WoodsHole,112: 185-201. - 1960. Histological studies on the digestive system of a starfish,Henricia,with notes on Tie- demann's pouches in starfishes. Ibid., 119: 371-398. BLEGVAD, H. 1914. Food and conditions of nourishment among the communities of invertebrate animals found on or in the sea bottom in Danish waters. Rep. Dan. biol. Stn, 22: 41-78. BULL, H. O. 1934. Aquarium observations on the rate of growthand enemiesof the common starfishAsterias rubens.Rep. Dove mar. Lab., 3. Ser.,2: 60-65. CAREY, A. G., Jr. 1965. Preliminary studies of animal-sediment interrelationships off the central Oregon coast. In Ocean Science and Ocean Engineering 1965. (Marine Technology Society). 1: 100-110. 1968. Zinc-65 in echinoderms and sediments in the marine environment off Oregon. Proc. Second Nat. Radioecology Symp. USAEC Conf. 6705-03, pp. 380-388. - 1972. 65Zn in benthic invertebrates off the Oregon coast, pp. 833-842.InA.T. Pruter & D. L.Alverson (eds.): Columbia River Estuary and Adjacent Ocean Waters: Bioenviron- mental Studies. Univ. Washington Press, Seattle, Washington. -Benthic infaunal abundance on two abyssal plains in the Northeast Pacific Ocean. (unpublished manuscript). CHRISTENSEN, A. M. 1970. Feeding biology of the sea-starAstropecten irregularisPennant. Ophelia, 8: 1-134. EKMAN, S. 1953. Zoogeography of the Sea. Sidgwick and Jackson. London. 417 pp. FEDER, H. & A. M. CHRISTENSEN. 1966. Aspects of asteroid biology.In R.A. Boolootian (ed.): Physiology of Echinodermata. pp. 87-127. Interscience, New York. FENCHEL, T. 1965. Feeding biology of the sea starLuidia sarsiDiiben & Koren. Ophelia, 2: 223-236. FISHER, W. K. 1911. Asteroidea of the North Pacific and adjacent waters. Part 1. Bull. U. S. natn. Mus., 76(1): 419 pp. - 1928. Asteroidea of the North Pacific and adjacent waters. Ibid., 76(11): 245 pp. GISLEN, T. 1924. Echinoderm Studies. Zoo]. Bidr. Upps., 9: 11-316. GRIEG, J. A. 1913. Bidrag til Kundskapen om Hardangerfjordens Fauna. Bergens Mus. Aarb., 1913, no. 1: 147 pp. HOPKINS, T. S. & G. F. CROZIER. 1966. Observations on the asteroid echinoderm fauna occurring in the shallow water of southern California (Intertidal to 60meters). Bull.Sth. Calif. Acad. Sci., 65(3): 129-145. HYMAN, L. H. 1955. The Invertebrates: Echinodermata. The CoelomateBilateria.Vol. IV. Me Graw-Hill, New York. 763 pp. MADSEN, F. J. 1956. Echinoidea, Asteroidea, and Ophiuroidea from depths exceeding 6000 meters. Galathea Rep., 2: 23-32. - 1961. On the zoogeography and origin of the abyssal fauna in view of the knowledge of the Porcellanasteridae. Ibid., 4: 177-218. MACGINITIE, G. E. & N. MACGINITIE. 1949. Natural History of Marine Animals. McGraw- Hill, New York. 473 pp. 143

FOOD SOURCES OF ASTEROIDS 47

MAUZEY, K. P., C. BIRKELAND & P. K. DAYTON. 1968. Feeding behavior of asteroids and escape responses of their prey in the Puget Sound region. Ecology, 49: 603-619. MCCAULEY, J. E. 1972. A preliminary checklist of selected invertebrates from otter trawl and dredge collections off Oregon, pp. 409-421.InA.T. Pruter & D.L.Alverson (eds.): Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. Univ. Washington Press, Seattle, Washington. MORTENSEN, TH. 1927. Handbook of the Echinoderms of the British Isles. Oxford Univ. Press, Oxford. 471 pp. PEARSE, J. S. 1965. Reproductive periodicities in several contrasting populations ofOdonaster validusKoehler, a common Antarctic asteroid. Biology of Antarctic Seas. II. Antarctic Res. Ser., Washington, 5: 39-85. SANDERS, H. L. & R. R. HESSLER. 1969. Ecology of the deep-sea benthos. Science, N.Y., 163: 1419-1424. SOKOLOVA, M. N. 1957. The feeding of some carnivorous deep-sea benthic invertebrates of the far eastern seas and the northwest Pacific Ocean: Trudy Inst. Okeanol., 20: 279-301. (Cited from the translation published 1959 by Amer. Inst. Biol. Sci., pp. 227-244). - 1958. The nutrition of some detritus-feeding invertebrates of the deep-water benthos. (In Russian) Trudy Inst. Okeanol., 27: 123-153. - 1959. On the distribution of deep-water bottom animals in relation to their feeding habits and the character of sedimentation. Deep-Sea Res., 6: 1-4. - 1966. La structure trophique du benthos de l'abyssal. Moscow, 1966. Second Internat. Ocea- nog. Congress, Abstracts of Papers: pp. 345-346. THORSON, G. 1957. Bottom communities. In J. W. Hedgpeth (ed.): Treatise on marine ecology and paleoecology (Mem. geol. Soc. Am., 67), vol. 1, Ecology, pp. 461-534. 144

RLO-2227-T12-19 submitted to Crustaceana

A REDESCRIPTION OF PETALIDIUM SUSPIRIOSUM BURKENROAD, 1937 (DECAPODA:NATANTIA)

RobertA.Wasmer Department of Zoology Oregon State University Corvallis, Oregon U.S.A.

Prior to 1937,the genus Petalidium, a poorly known genus in the penaeid shrimpfamily Sergestidae,was consideredto differfrom the genus Sergestesin the smallernumber ofgills above the fourthpereiopods, in the lesserdegree oframificationof its gills,and in thebifurcation of the processusventralisof its petasma(Hansen,1922;Burkenroad, 1937). In 1937,Burkenroad described PetaZidiurn suspiriosum from two female specimenstaken in theeastern Pacific offMexico. These specimens differed from otherspecies of PetaZidium in having the same number of gills above the fourthpereiopods as does Sergestes, and in having considerably more branched gills.

Over 1000specimensofPetaZidium suspirioswn Burkenroad are among the mesopelagicdecapod crustaceanscollectedsince 1961 by the Department of Oceanographyof OregonState Universityin the northeastern Pacific Ocean. Sincethis speciesis only knownfrom Burkenroad's two typespecimens,and since the male iscompletely unknown,it seems desirable to redescribe the species from the material now available. The collection of the specimens by vessels of the OSU Department of Oceanography was supported byU.S.Atomic Energy Commission Contracts AT(45-1)-1726 andAT(45-1)-1750, RLO-2227-T12-19. I wish to thank Dr. WilliamG.Pearcy of the Department of Oceanography andDr.Ivan Pratt of the Department of Zoology for their suggestions and helpful criticisms made in reviewing the manuscript.

Petalidium Bate, 1881 The followingdiagnosis of the genus PetaZidium is based largely on Hansen(1922), with additionsmade to indicate the increase in the number of gills and their increased ramification inPetalidium suspiriosum: 145

Sergestidae with first 3 pairs of pereiopods elongate; 1st pereiopod without proper chela; 2nd and 3rd pereiopods with very small chelae, 4th and 5th pereiopods with 6 segments, the dactyli being absent, 5th much shorter than 4th, both natatory. First maxilla with palp; 2nd maxilla with 2 lobes; 1st maxilliped with segmented palp. Branchial lamellae as well as arthrobranchs present; arthrobranchs of up to 13 rami, with as many as 12 lamellae per rami; lamellae relatively large and independent. Number of arthrobranchs above 4th pereiopods variable:none, a single rudimentary, or two may be present; none above 5th. Petasma with processus ventralis forked.

PetaZidium suspiriosumBurkenroad, 1937

PetaZidium suspiriosumBurkenroad, 1937: 325, figs. 8-12. P. suspiriosum- Pearcy & Forss, 1966: 1136, 1137, 1140.

Type. - Burkenroad's (1937) type specimens of P.suspiriosumwere collected in the eastern Pacific off Mexico, 20°36'N 115°07'W, and are deposited in the collections of the Department of Tropical Research of the New York Zoological Society, Catalogue number 361,043.

Diagnosis. - A well-developed anterior arthrobranch of nine rami, with up to six lamellae per rami, and a small posterior arthrobranch of three rami bearing several lamellae each above fourth pereiopod. An anterior arthrobranch, which may have up to thirteen rami, withup to twelve lamellae per rami, and a posterior arthrobranchial lamella above third maxilliped and first three pereiopods. Posterior, secondarybranch of lobus armatus of petasma reaching beyond middle of primary branch, armed withup to six crochets; external branch of processus ventralis reaching beyond external branch, and almost as far as internal branch, of lobus terminalis; internal branch of lobus terminalis armed withone or two terminal crochets. Small hepatic spine present. Ocular peduncle with two tubercles.

Description. - The followingis based onnew material of P. sus- piriosumfrom the northeastern Pacific Ocean, deposited in the U.S. National Museum, Washington, D.C., Catalog Nos. 139318-139323.

Integument thin and membraneous; most specimens seen are damaged, with distal articles of antennular peduncles, tips of antennal scales, third maxillipeds, pereiopods, and distal portion of uropods missing.

Male. Rostrum (fig. 1) very short and moderately high, superior margin convex; tip rounded, without dorsal spines. Carapace without dorsal carina. At about the level of hepatic spine, dorsal midline of carapace is elevated into a variously shaped hump, quite low in some specimens, but in others in form of a conspicuous, anteriorly directed 2

Figs. 1-4.Petalidium suspiriosum Burkenroad.1. carapace, lateral view, male (carapace length9.4 mm); 2. telson, distal portion, male (c.l. 9 mm); 3.left eye in dorsal view, male (c.l. 9 mm);4.. right inferior antellal flagellum, medial view,male (c.l. 9.5 mm); Scale for fig. 1= 3 mm; for fig. 2 = 0.5 mm; for figs. 3,4= 1 mm. 147 sub-conical projection. Cervical sulcus very distinct, continued across dorsum, accompanied on carapace sides by cervical carina. Branchiocardiac carina and antennal carina well-developed. Weakly developed inferior carina and sulcus extending obliquely downward and backward across anterior portion of branchial region. Supraorbital spine absent. Minute hepatic spine present; may occasionally be missing on one or both sides, in which case a distinct hepatic prominence remains.

Abdominal somites not dorsally carinate. Sixth somite as long as fourth and fifth combined, terminating in a minute tooth. Telson slightly more than half as long as the external margin of uro- podalexopodite,narrowing abruptly to a small terminalspine,often flanked by a pair of fixed lateral teeth (fig.2). Uropodal exopodite slender, a littleless than five times as long as wide; less than distal one-seventh of external bordersetose,with a well-developed tooth between setose and non-setoseportions. Subcuticular tissue of telson and uropods reticulate in appearance. Cornea of eyes (fig. 3) wider than long, about same width as peduncle, dark brown in color.Well-developed tubercle at base of cornea on inner side of peduncle, and a smaller projection more proximally on peduncle. First article of antennular peduncle less than twice as long as wide, with stylocerite in form of a very small acute tooth near middle of outer margin; slightly longer than outer margin of second and third together. Outer margin of third article slightly less than 1.5 times as long as wide, about samelength as outer margin of second.

Inferior antennular flagellum (fig. 4)as long astwo distal articles of antennular peduncle, divided into peduncle of four articles and a flagellum. Upper prolongation of third article very long, reaching to the third article of flagellum; distal portion with a strong spine. Inferior margin of prolongation somewhat concave, armed with 16 large, distally blunt and spat- ulate spines. Superior margin of fourth article deeply concave, terminating in a high, rounded protuberance; concave portion armed withnine spines similar to those on prolongation of third article,but somewhat smaller. Flagellum of five articles; last article with single large spine.

Antennal scale three times longer than greatest width, narrowing distally with both margins convex; extending to distal half of third article of antennular peduncle. Apex of lamella extending beyond vestige of small distal tooth on external margin. Subcuticular tissue of scale reticulate in appearance.

Antennal flagellum extremely long. In only specimen seen with attached flagellum, the flagellum is about ten times carapace length, extending approximately 97 mm compared with a carapace length of about 9.5mm, but 148 is obviously longer in life as a distal portion is missing. Flagellum divided into a proximal non-setose and a distal setose portion by region in which the annuli form a double bend; annuli of distal portion bear paired plumed setae which arch upwards toward each other, between which, at various intervals of annuli, single straight plumed setae project upward.

Mandible with palp of three articles; second article one and one-half times as long as third, which is slender.

First maxilla has two strong brown spinesdistally on palp.

Second maxilla has endite of two lobes; proximal lobe relatively narrow and setose, distal lobe distinctly divided by an incision into two parts, proximal part narrow; endopodite armed distally with spines.

Endopodite of first maxilliped consists of three articles, reaching beyond exopodite; the distal lobe of endite two and one-half times longer than wide, proximal lobe subdivided by a suture into two parts; epipodite large and well-developed.

Second maxilliped thickly covered with golden-brown setae. Dactylus slightly more than half as long as propodus, narrow proximally, widening towards middle.

Third maxilliped reaching to distal end of antennular peduncle. Neither propodus nor dactylus divided into subsegments.

First pereiopod non-chelate, reaching to distal end of first article of antennular peduncle. Carpus has four long setae in middle of inner margin; a comb-like arrangement of setae distally on carpus and proximally on propodus, in opposition across the articulation between the segments.

Fourth pereiopod reaching just beyond distal end of first article of antennular peduncle. Carpus and propodussetose ononly one margin.

Fifth periopod short, reaching to proximal fifth of merus of fourth pereiopod. Carpus and propodus setose on only one margin.

First maxilliped with exopodite and epipodite, no branchiae; podobranch and arthrobranchial lamella (= rudimentary arthrobranch) above second maxilliped; an anterior arthrobranch, which may have up to thirteen rami, with up to twelve lamellae per rami, and a posterior arthrobranchial lamella above third maxilliped and first three pereiopods; a well-developed anterior arthrobranch of nine rami, with up to six lamellae per rami, and a small posterior arthrobranch of three rami bearing several lamellae each above fourth pereiopod; no branchiae above fifth pereiopod. 1'9

The nomenclature for describing the petasma is that used by Hansen (1919, 1922). Petasma of adult (figs. 5-6) with pars externa an oblong plate about four times as long as wide, not divided into lamina externa and processus uncifer; distal border obliquely truncate, with sharp oblique tooth. Processus basalis moderately long, posteriorly directed, distally blunt. Pars astringens a flat, folded, rectangular plate attached to proximal part of inner margin of pars media; inner margin straight, with row of small cincinnuli. Pars media long, wide at base, narrowing gradually.

Processus ventralis long, somewhat curved, directed anteriorly and a little externally; proximally slender, thickening towards middle, where it is divided into two branches; internal branch less than half as long as external branch, terminating in one small crochet (retractable hook); external branch curved medially, narrowing distally, extending almost as far as internal branch of lobus terminalis, with 12 or 13 oblique, triangular crochets on its internal surface.

Lobus armatus divided into two branches; primary branch very thick proximally, almost conical, directed externally and to posterior, with a series of up to 11 large triangular crochets on distal part of its anterior margin; smaller secondary branch originates on posterior surface of primary branch, reaching to beyond middle of primary branch, armed with up to six crochets, including two distally.

Lobus connectens in the form of a low rounded protuberance, visible on anterior side of petasma, between lobus armatus and lobus inermis, without hooks.

Lobus inermis conical, without hooks.

Lobus terminalis originates behind lobus connectens and lobus inermis, long, extending past processus ventralis; divided into two branches a little before its middle; external branch slim, as long as undivided portion of lobe; internal branch longer than external branch, wide, with one or two terminal crochets.

Appendix masculina (fig. 7) on second pleopod terminates in 11 spines, with three long spines on distal portion of internal surface.

Female. Females usually larger than males taken in same haul. Rostrum differs from that of males by having a pointed tip and usually a rudimentary dorsal tooth. Genital area of female (fig. 8) lacks an operculum; posterior lip of receptacular atrium incised medially, overlying anterior lip. Large triangular lamella arising from articulation of coxa of third pereiopod extends about half way to incised region of posterior lip. Postero-median margin of coxa of third pereiopod well-developed, in form of elongate, blunt projection, which fits into a hollow on oblique ridge at posterior edge of plate between bases of third pereiopods. 150

Figs. 5-8. PetaZidium suspiriosumBurkenroad. 5. right petasma, anterior view (carapace length 9.5 mm); 6. distalportion of right petasma, posterior view (c.l. 9.5 mm); 7. appendix masculina,medial view (c.1.9.5mm); 8. genital area, female, showingcoxae of third periopods (c.l. 12.5 mm). Scale = 1 mm. a. = pars astringens,e. = pars externa, la. = lobus armatus, ic. = lobus connectens, li.= lobus inermis, lt. = lobus terminalis, m. _ pars media, pb. = processus basalis,pv. = processus ventralis. 151

Size. - Due to the fragile nature of the integumentofP. suspiriosum and the resulting damage tomany of the specimens during capture, it is' often difficult to measurecarapace lengths accurately. Therefore, in many cases, no attempt was made to measurecarapace lengths. In those specimens which were measured, carapace lengthsof males ranged from 7 mm to 11 mm; specimens with a carapace length of less than8 mm invariably had a petasma with incompletely developed lobesand crochets. Carapace lengths of females ranged from 8 mm to 14 mm.

Colorinformalin. -Fresh materialcarmine,abdominal somiteslighter; stomach and esophagus deepcarmine. Older material faded to which except for stomach, esophagus,andmouthparts,which tend to retaintheircarmine color.

Distribution. - Northeastern Pacific Ocean. Specimens in the OSU Department of oceanography collections have beentaken as far north as 51°40.9'N 138°30.3'W andas far south as 34°23'N 140°46'W (Burkenroad, 1937). Although specimens have been taken by midwater trawlswithin the upper 200 m,PetaZidium suspiriosumappears to be mainly mesopelagic in distribution. The results of collections madeover a period of three years from the upper 1500 m of water off thecentral Oregon coast (44°39'N 125° 15'W) show P.suspiriosumto be most abundant in the depth range of 200- 1000 m (Pearcy and Forss, 1966).

Remarks. - As was pointed out by Burkenroad (1937),PetaZidium suspiriosumdiffers from other members of thegenus as follows in the structure of the gills: rami of the anterior arthrobranchs above the third maxilliped and first three pereiopodsmay have as many as twelve lamellae, instead of generally fiveor six (Hansen, 1922); and two arthrobranchs are present above the fourth pereiopod (the anteriorone large and well-developed)-, instead of a single rudimentaryone as inP. obesumKroyer, 1859 (Hansen, 1922) or none as inP. foZiaeeumBate, (Bate, 1888; Hansen, 1903; Illig, 1914).

Otherwise,PetaZidium suspiriosumdiffers from P.obesum(as described and illustrated by Hansen, 1922)as follows: integument not reticulate except in telson, uropods, and antennal scales; cervicalsulcus distinct; hepatic spines present; a second tuberclepresent proximal to well-developed one distally on ocular peduncle; and genitalarea of female without operculum over lips of receptacular atrium.

The petasma ofPetaZidium suspiriosumdiffers from that of P.obeswn chiefly in the size and armature ofthe posterior, secondary branch of the lobus armatus, reaching beyond the middleof primary branch and armed with up to six crochets in P.suspiriosum,instead of being much smaller than the primary branch and armed with twocrochets as in P.obesum;external branch of processus ventralis reachingbeyond external branch, and almost as far as internal branch, of lobus terminalis; and internalbranch of lobus terminalis armed withone or two terminal crochets, instead of three on its distal external face as inP. obesum. 152

Illig (1927, p. 283) published a rather poor figure of the petasma of a specimen of PetaZidium, identified by him as P.foliaceum,taken in the South Atlantic not far north of the previous records for the species; according to Burkenroad (1937), this figure agrees with Hansen's (1922) figures of the petasma of P.obeswn(it also bears some resemblance to the petasma of P.suspiriosumas herein described). PetaZidium obesum otherwise has. been taken only in the North Atlantic, having been recorded as far north as 40°15'N 56°25'W (Hansen, 1922).

PetaZidium suspiriosumdiffers from P.foliaceum Bate mostclearly in the number of gills above the fourth pereiopod, as pointed out above. The information concerning the nature of the cervical sulcus, the presence or absence of supraorbital and hepatic spines and of a second tubercle on the ocular peduncle in P.foliaceumis incomplete and sometimes contradictory. Burkenroad (1937) draws together information on these features of P.foZiacewn from descriptions of the species by Bate (1888), Hansen (1903), Illig (1914, 1927), and Stebbing (1914).According to Hale (1941), the five specimens of P.foZiaceumtaken by the British Australian and New Zealand Antarctic Research Expedition agree closely with the "Challenger" specimens (Bate, 1888; Hansen, 1903). The cervical sulcus is well-marked; there are no supraorbital spines, but minute and easily overlooked hepatic spines are present; and the ocular peduncle has two tubercles in all the specimens. The genital area of the female appears to be much like that of P.suspiriosum.

Figures of the petasma of P. foliaceum(Illig, 1914; Stabbing, 1914; and Hale, 1941) are in close. agreement, and differ from that of P.suspiriosumas follows: the posterior, secondary branch of lobus armatus is larger and more heavily armed that the primary branch; external branch of processus ventralis reaches about as far as external branch of lobus terminalis; and internal branch of lobus terminalis has a series of large crochets on its external face. PetaZidium foliaceum appears to be subantarctic in distribution (Hansen, 1920),having been reported from the Pacific and Atlantic sectors of the Antarctic Ocean. PetaZidium is closely allied to Sergestes (Hansen, 1922, p. 189; Burkenroad, 1945, p. 577).Prior to the description of P. suspiriosum, PetaZidium was considered to differ from Sergestee in the lesser degree of ramification of its gills, in the smaller number of gills above the fourth pereiopods, and in the bifurcation of the processus ventralis of its petasma (Hansen, 1922; Burkenroad, 1937).The differences in gills amounted to the following: The arthrobranchs in PetaZidium had a much lower number of rami, a much lower number of lamellae per rami, with the lamellae much larger, curved upward, and looking much more independent than those in Sergestes; and in Petalidium there were no arthrobranchs above the fourth pereiopods (P. foliacewn) or a single rudimentary one (P. obesum), while in Sergestes there are always two arthrobranchs above the fourth pereiopods. 153

As has been pointed out by Burkenroad (1937), the condition of the gills inP. suspiriosum differs fromthat previouslydescribedfor the genus, in that the gills are considerablymore ramified(through stillless so than inSergestes),and that thereare the same number abovethe fourth pereiopod as in Sergestes. Besides the biramous condition of the processus ventralis of the petasma of Petalidium,Hansen (1922, p. 190) methions that the petasma differs from that inSergestesin that thepars externa is not divided into a lamina externa and processusuncifer,and that the pars astringens is reduced in regards tosize. The petasma of P. suspirioswn, as hereindescribed,agrees in these characters with the other petasmata described for the genus.

Other characters given by Hansen (1922) by whichPetalidiumdiffers fromSergestesare that the fringed portion of the outer margin of the uropodal exopodite is much shorter inPetaZidiumthan in anySergestes,that the telson is quite short andwide,without dorsalspines,and that the inferior antennal flagellum of themale,which is modified into a clasping organ, is very much different inPetaZidiumthan inSergestes.However, due to the often damaged condition of specimens of Petalidium, these last differences are of limited usefulness in distinguishing between the two genera. PetaZidium apparently differs further from the majority of species of Sergestes in lacking luminous organs of the types known to be found in Sergestes,Three different types of luminous organs occur in Sergestes, although no species has more than one type and some have none at all (Yaldwyn, 1957). Yaldwyn (1957), divided thegenus Sergestesinto the subgenera Sergestes and Sergia by the presence or absence of organs of Pesta (luminescent modifications of the gastrohepaticgland). Within the subgenus Sergiaisa group of species lacking dermal photophores and having membraneous integuments (the "S. japonicas" group of Yaldwyn). According to Burkenroad(1937),P. suspiriosum shows "considerable super- ficial resemblance" toSergestes mollis (=Sergestes (Sergia)japonicus) andSergestes (Sergia) inous,both members of the"S. japonicus"group; P. suspiriosum can be distinguished from these species (Burkenroad, 1937) by differences such as the appearance of its gills, by the presence of an hepaticspine,by the absence of more than one pair of lateral spines on itstelson,and by its shorter but larger eyes with distomedially tuberculate peduncle.

The developmentofPetalidium is known with certainty only as far back as the early mastigopus(post-larval) stages,and in view of the general near-identy of the mastigopus and adults of Sergestes and PetaZidium (Burkenroad, 1945,p.566-567),it is unlikely that earlier larvae will differ greatly from those of Sergestes.A series of sergestid larval stages were attributedtoPetalidium by Gurney (1924), but later were regarded as belongingtoSicyonella Borradaile (Gurney andLebour, 1940).These same 154

larvae (referred to now as "Gurney's larvae")were predicted by Burkenroad (1945) to belong to an as yet undiscovered seventhgenus of Sergestidae, probably closely related toPeisos.

The presence in Petalidium suspiriosumof an antennal flagellum which is divided into a proximal and a distal portion bya section of annuli forming a double bend, with the distal section being heavily and characteristically setose, makes it possible to addPetaZidium suspiriosumto the list of species (Foxton, 1969) which have such antennae. Foxton records such antennae from species of three genera, including the penaeidsFunehaZiaandGennadas,and the sergestidSergestes; to thesehe adds the sergistidAceteson the basis of evidence presented by Burkenroad (1934)4 of these, only the species of Funchalialack the kink or double bend in the flagellum, although a point of flexure is present between the proximal non-setose and the distal setose sections (Foxton, 1969).

The hump or projection in the anterior dorsal midline ofPetalidium suspiriosumagrees very well with Hansen's (1921) description quoted in Hansen, 1922, p. 20) of the so-called "dorsal organ" found inmany Sergestes larvae and adults. Although its function is disputed, the "dorsal organ" can be traced in the embryos, larvae, or adults of most groups of Crustacea (Gurney, 1942). This is apparently the first record of its occurance in PetaZidium.

ZUSAMMENFASSUNG

PetaZidium suspiriosumBurkenroad wird wiederbeschrieben an Hand von Material, das im nordostlichen Pazifik westlichvon Oregon gesammelt wurde. Die Unterschiede zwischenPetalidium suspiriosumand den anderen Arten dieser Gattung, and zwischen den GattungenPetalidiumandSergesteswerden diskutiert. Die Antennengeissel vonP. suspiriosumist gleich der Geissel wie sie bei eingen Arten der GattungenFunchalia, Gennadas, Sergestesand Acetesvorkommt. Ein "Ruckenorgan", wie es in der GattungSergestes vorkommt, ist ebenfalls bemerkt.

REFERENCES

BATE, C. S., 1881. On the Penaeidea. Ann. Mag. Nat., ser. 5, 8:169-196, pls. 11, 12.

1888. Report on the Crustacea Macrura collected by H.M.S. Challenger duringthe years 1873-76. Rep. Voy. Challenger,(Zool.) 24: 1-942, text- figs.1-76, pls. 1-150. 153

BURKENROAD, M. D., 1934. The Penaeideaof Louisianawith a discussion of their world relationships. Bull. Amer.Mus.nat.Hist.,68(2): 61-143, figs. 1-15.

1937. The Templeton Crocker Expedition. XII.Sergestidae (Crustacea Decapoda) from the Lower California regionwith descriptions of two new species and some remarkson the organs of Pesta in Sergestes. Zoologica, N. Y., 22(4): 315-329, figs. 1-12.

1945. A new sergestid shrimp (Peisos Petrunkevitchi,N.Gen., N. Sp.), with remarks on its relationship.Trans. Conn. Acad. Arts Sci., 36: 553-591, pls. 1,2. _

FOXTON, P., 1969. The morphology of theantennal flagellum of certain of the Penaeidea (Decapoda, Natantia).Crustaceana, 16(2): 33-42, figs. 1-3, pl. 1.

GURNEY, R., 1924. Crustacea. IX. Decapod larvae. British Antacrctic (Terra Nova) Exped. 1910, Nat. Hist.Rep., Zool., 8(2): 37-202, figs. 1-78.

1942. Larvae of decapod Crustacea: 1-306. (The Ray Society,London).

GURNEY, R., & M. LEBOUR, 1940. Larvae of decapod Crustacea Part VI. The genus Sergestes. Discovery Rep., 20: 1-68, figs. 1-56.

HALE, H. M., 1941. Decapod Crustacea. Brit. Aust. New Zeal.Antarct. Res. Exped. Rep., (B)4: 257-286, figs. 1-16, pl. 3.

HANSEN, H. J., 1903. On the crustaceans of the genera Petalidium and Sergestes from the Challenger, withan account of luminous organs in Sergestes challengeri,n. sp. Proc. zool. Soc. London, 1903(1): 52-79; pls. 11, 12.

1919. The Sergestidae of the SibogaExpedition.Siboga Exped., 38: 1-65, text-figs. 1-14, pls. 1-15.

1920. Les Sergestidae des Expeditions du Travailleuret du Talisman. Bull. Mus. natn. Hist. nat., Paris, 1920(6):477-483.

1921. On the postembryonic occurance of the median "dorsalorgan" in the Crustacea Malacostraca. Studies on Arthropoda I. Copenhagen.

1922. Crustaces Decapodes (Sergestides)provenant des campagnes des Yachts Hirondelle et Princesse-Alice (1885-1915).R6s. Camp. sci. Monaco, 64: 1-232, pls. 1-11.

ILLIG, G., 1914. Die Dekapoden der Deutschen Siidpolar-Expedition1901-1903. II. Die Sergestiden, Deutsch. SUdpolar-Exped., (Zool.)7: 349-376, figs, 1-38.

* Not Seen 156

1927. Die Sergestiden der DeutschenTiefsee-Expedition. 3. Natantia, TeilB. Wiss. Ergebn.ValdiviaExped.,23: 279-354,figs.1-131.

*KROYER,H.,1859. Forsog til en Monographisk Fremstilling of Kraebsdyrs- laegtenSergestes.Med Bemaerkninger om Dekapodernes Horeredskaber. K. Danske Vidensk.Selskskr.,(5)4: 217-302, pls. 1-5.

PEARCY,W. G., & C.FORSS, 1966. Depth distribution of oceanic shrimps (Decapoda;Natantia) offOregon.J.Fish.Res.Bd. Canada,28: 1135-1143.

STEBBING,T. R. R.,1914. Stalk-eyed Crustacea Malacostraca of the Scottish National Antarctic Expedition.Trans.Roy.Soc. Edinburg,50(2): 253-307, pls. 23-32.

YALDWYN,J. C., 1957. Deep-water Crustacea of the genus Sergestes (Decapoda, Natantia)from CookStrait,NewZealand. Zool. Publ.Victoria Univ. New Zealand,22:1-27, figs. 1-19.

* Not Seen RLO-2227-T12-20 15 1

Reprinted from LIMNOLOGY AND OCEANOGRAPHY Vol. 17, No. 6, November 1972 pp. 833-839 Made in the United States of America A TECHNIQUE FOR THE ESTIMATION OF INDICES OF REFRACTION OF MARINE PHYTOPLANKTERS' Kendall L. Carder2, Richard D. Tomlinson, and George F. Beardsley, Jr.' School of Oceanography, Oregon State University, Corvallis97331

ABSTRACT Measurements of light scattering and particlesizedistribution were performed on a growing culture of the unicellular phytoplankterIsochrysis galbana.The culture represented a narrow, polydisperse distribution of nearly spherical particles.Assuming the culture to be homogeneous in refractive index, a technique was developed that provided an estimate of the index of refraction of the culture for the wavelengths 5,460 A and 5,780 A. The relative index ofrefraction(relative to that of seawater) of the culture varied with time from 1.026 to 1.036 over a 12-day sampling period and seemed to be related to changes in the surface area : volume ratio of the cells.The indices of refraction determined using green light (5,460 A) corresponded closely with those found using yellow (5,780 A), with slight fluctuations perhaps due to changes in cell pigmentation ratios.

INTRODUCTION suggested that more than half of the open- ocean particles are smaller than 1 µ in Light propagation in the sea is a func- This means that they average tionof both thequantity and opticaldiameter. quality of suspended particulates in theroughly one wavelength in radius. Refrac- water. Particle quantity is quite easily de-tion techniquesare inadequateinthis terminable by filtration or electronic count-small size range even if the particles are ing (e.g. Coulter counter) techniques, butregular in shape. As a result, only ranges optical quality has been more difficult toof the relative index of refraction of oce- measure. For the purposes of light propa-anic particulates have been reported, usu- gation studies, the optical quality of par-ally based on the indices of the component ticulates can be considered to be limitedsubstances making up the particles. Burt to particle size, shape, and index of refrac-(1956) and Jerlov (1968) indicated ranges tion (relative to seawater). The sizes andof the relative index of refraction running shape of oceanic particulates have beenfrom 1.0 to 1.4, with that of bacteria near described by means of a number of tech-the lower end and hard substances (SiO2, niques(e.g.optical and electron micro-CaCO3, etc.) near the upper end. Bryant scopes), but their indicesof refractionetal.(1969) found quantitative agree- have been virtually unreported. ment of theoretical optical cross-sections The determination of the indices of re-predicted by Mie theory and anomalous diffraction theory with laboratory observa- fractionof oceanic particulatesisvery tions fromEscherichia colicells, spinach difficult because of their small size. Ocha-chloroplasts, and yeast cells.Their tech- kovsky (1966) and Beardsley et al. (1970)niques are also applicable to phytoplankton 1 This research was supported by the followingsuspensions but require extensive instru- sources:Office of Naval Research Contract ONR mentation and appear to be quite time N00014-67-A-0369-0007 under project NR083-102; consuming. Atomic Energy Commission Contract No. At (45- Measurements of the indices of refrac- 1)2227, Task Agreement 12 (RLO-2227-T12-20); and Federal Water Pollution Control Administra- tion of various species of phytoplankton tion Grant No. WP 111-04. remain unreported, even though phyto- 2 Presentaddress: MarineScienceInstitute, plankters represent the primary particle Universityof South Florida,830 FirstStreetproducers in the open ocean. Such infor- South, St. Petersburg33701. Deceased. mationisnecessary for a better under- LIMNOLOGY AND OCEANOGRAPHY 833 NOVEMBER 1972, V. 17(6) 158

834 CARDER, TOMLINSON, AND BEARDSLEY

Jerlov (1953) noticed that for seawater the radiant flux scattered at 0 = 45° was proportional to the totalflux scattered. Deirmendjian (1963) showed theoretically the proportionality constant of this relationship around 0 = 400 to be 0.1 regardless of the size distribution function. He 3 4 5 6 7 e+12 n d/ ,\)(M-I) also suggested that the measured intensity Ftc. 1.Scattering efficiency factor (for m near of light scattered around 0 = 40is directly unity). proportional to the concentration of particles, provided that the distribution is con- standing of the wide variability in thetinuous with a real and constant index of scattering per unit geometrical cross-sec-refraction. tion reported by Pak et al. (1970), Carder et al. (1971), and Schlemmer and Carder TECHNIQUES (1972). We have verified (Beardsley et al. 1970) BACKGROUND that the volume scattering function, 8(0), for 0 = 45° can be approximated for sea- Mie (1908) demonstrated that the lightwater by scattered by a dielectric sphere placed in a plane-parallel beam of monochromatic f3(45°) = k 1 (K.,i) (Si), (2) light is a function of its size and index of refraction (relative to that of the medium)where K8,is the scattering efficiency fac- for a given wavelength of light.Sincetor of the ith particle, SI is the cross-sec- the scattering due to a dilute, homogene-tional area of the ith scattering particle, ous concentration of spheres isadditive,n is total number of scatterers, and k is Mie theory isquite useful for studyingan empirical constant of proportionality. scattering in the oceans. If the particles of interest have a fairly Van de Hulst (1946) provided some sim-narrow size distribution, equation 1 can be plifications to the Mie theory for particleswritten as having a relative index of refraction near /3(45°) = kK*8nS*, (3) 1.0. He showed that the extinction efficiency factorQextcould be expressed as where K*8 and S* are mean values.Solv- ing equation 3 for K*8 produces Qext=2+ (4/p2) (1- eos p) - (4/p) sin p, (1) ,8(450) K*8 _ (4) wherep= (2ird/A) (m-1), dis the par- knS* ticle diameter (0 < d < oo ), x is the wave-Since for optically large particles K*8 = 2 length of light in the medium (0 < A < oo ) (van de Hulst 1946; Jerlov 1968), k can and m is the index of refraction of thebe determined experimentally using poly- particlerelativetothemedium.Fordisperse suspensions of large latex spheres. nonabsorbing particles the scattering effi- Since K8 is not monotonic over its entire ciency factor K8 isequal toQext,andrange of values and approaches the limit equation 1 then also represents the scat-2.0 for large values of p, it is important tering efficiency factor. Figure 1 illustratesto select particles falling in the monotonic thatK8is monotonic only for small valuesrange of K*8 to obtain a one-to-one corre- of p. To attempt to use values of Ke, A,spondence between K*,, and p. anddto definemfor larger values ofp Because phytoplankters are the primary would be more difficult, but could beproducers of oceanic particulates, a knowl- accomplished by varying A sufficiently foredge of their indices of refractionisa recognition of the curve shape. necessary step in describing the submarine 1°,9

REFRACTION INDICES OF MARINE PHYTOPLANKTERS 835

light field.Phytoplankton cultures repre-that of Davis and Ukeles (1961).The sent relatively narrow polydisperse distri-algae were grown under conditionsof butions of particles with generallynarrowconstant irradiance(1.7 mW/cm2) and ranges of refractive index; both of theseaeration. characteristics are essential for the application of techniques developed above. SAMPLING PROCEDURES The unarmored phytoplankterIsochry- Measurements of light scattering and sis galbanawas selected for this studyparticle size distribution were made on because of its optical size, relatively highsmall samples.Optical sampling was be- transparency, and nearly spherical shape.gun during the exponential growth phase Its values of K*8 fall on the monotonicof the culture. Each sample was diluted portion of Fig.1 (p < 2), so that thewith autoclaved, filtered seawater to re- value of p corresponding to a given K*8duce coincidence levelsof the Coulter is unique. counter to below 1% of the total number of particles. ALGAL CULTURE METHODS Measurements of 8(45°) were taken Isochrysis galbana(class Chrysophyceae,with a Brice-Phoenix light scattering pho- division Crysophyta) is unicellular, lies intometer(describedbySpilhaus1965; the nannoplankton size range (2-20 µ)onBeardsley 1966; Pak 1970) ; n and S* were the scale given by Dussart(cited inShel-determined with a Coulter counter model don and Parsons 1967a), and has beenA (described by Sheldon and Parsons thoroughly described by Parke(1949).1967b) and k empirically from the scatter- The optical work accompanieda study of °5'Zn uptake and retention for which theing by two narrow polydisperse systems culturing techniques were designed. Theof large (10.5 and 21.0 p) latex spheres. initial inoculum was obtained in axenic culture from Dr. F. B. Taub. Subcultures RESULTS were grown in autoclaved, filtered seawa- Optical measurements were made over ter held at 16.00 -L 0.25C. The growth me-a period of 13 days. During this time the dium used before inoculation of the uptakeculture passed from a phase of rapid and retention experiments was similar togrowth into one of population maintenance

TABLE 1.Cell growth and light-scattering data for a batch culture ofIsochrysis galbana*

Elapsed time $(45°) X 10-2 (days) n D* 2(11)2 Dilution D*(g) (m-sr)-1 K, P (m-1) 0 14,918 1:20 4.65 22.41 3.19 0.969 1.47 0.0225 2.70 0.820 1.34 0.0259 2.50 0.759 1.28 0.0265 1 10,200 1:40 4.64 21.25 2.83 1.33 1.78 0.0269 2.49 1.17 1.62 0.0316 2.27 1.06 1.55 0.0315 6 9,964 1:40 4.28 19.39 2.87 1.51 1.93 0.0313 2.27 1.19 1.67 0.0344 2.00 1.05 1.54 0.0340 8 10,974 1:40 4.24 18.92 3.00 1.47 1.89 0.0312 2.46 1:20 1.67 0.0349 2.32 1.13 1.62 0.0358 12 9,902 1:40 4.24 19.00 2.34 1.26 1.77 0.0287 1.83 0.988 1.49 0.0313 1.77 0.955 1.45 0.0326

For each sampling period, the values of )9(45°), K,, p, and (rn - I) are listed in order of blue (4,360 A),green (5,460 A), and yellow (5,780 A) wavelengths. 160

836 CARDER, TOMLINSON, AND BEARDSLEY

3 -0 0 'A-77 -7 0 5000 2 Yellow Li ht W

i Night E 0 Green F- _j 0 Blue Light M qeo f

U t I I n __L I I I I I I I I I I I I I I 0 1 2 3 4 5 6 7 8 9 0 I( 0 2 4 6 8 10 12 DIAMETER (p) TIME (days) Fic.2. A typical cumulative distribution of Fic. 4.The effects of time and wavelength on Isochrysis galbana cells. the index of refraction of Isochrysis galbana.

and senescence. Figure 2 shows a typicalably selectively exclude some of the less cumulative distribution curve forI.gal-viable organisms. bana.It represents a relatively narrow Figure 4 shows a time sequence of the polydisperse distribution.Table 1 showsrelative indices of refraction of the culture pertinent growth and scattering parame-for blue, green, and yellow wavelengths. ters:n is counts/cm3 of diluted culture;All curves show a general increase with dilution represents the ratio of culture ali-time until the eighth day, when they start quot to diluent; and zero days of elapsedto decline. The "blue" curve values are time represents the time of the first light- smaller than those of the "green" and "yel- scattering measurements. low" curves. Figure 3 shows the time sequence of n, D*, and nD*3.Reproduction occurred DISCUSSION from 0 to 8 days elapsed time as evidenced An analysis of variance was performed by the increase in the total cell numberon the index of refraction data using the (n/dilution)and thedecreaseinthewavelengths as blocks and the days (time) mean cell diameter. The total cell volumeas treatments; F values of both the treat- as represented by nD*3 remained nearlyments and blocks were much greater than constant following an initial increase afterthe 1% confidence limits (Table 2). Con- 1 day.It is important to keep in mindsequently both wavelength and time pro- that these data represent only those orga-duced significant variations in the relative nisms in suspension. They do not includeindex of refraction. The error mean square those cells which settle out and thus prob-was quite small, and (EMS)1/3.06 < No was used as the relative experimental error, where 3.06 represents the mean value 44 7+44 of 100(m-1). The increase in refractive index in each

TABLE 2.Analysis of variance with wavelength as blocks and time (days) as treatments

Number

A MeaN Diameter Source df SS MS F

O Na3 Blocks 2 0.4691 0.2345 82.28* 24 1 I I I .-_ I 4 Treatments 4 1.513 0.3782 132.70+ 0 2 4 6 6 0 2 Error 8 0.0228 0.00285 TIME (days) Total 14 2.0049

Fic. 3.Effect of time on the growth of a * 1% confidence level :F2 (0.01) = 8.65. culture of Isochrysis galbana. t 1% confidence level:F4(0.01) = 7.01. 16:

REFRACTION INDICES OF MARINE PHYTOPLANKTERS 837 of the curves in Fig. 4 corresponds to theof refraction of cells when this nonabsorp- decrease in the mean cell diameter D*.tive technique is used. Due to the rela- Since the cell wall represents a relativelytively high absorption of blue light by refractive portion of the cell composition,chlorophyll, the indices of refraction de- this might have been expected, for therived using blue light are in error and smaller cells have a greater surfacearea : must be discarded. volume ratio. The increase in the index of refraction Table 2 also shows a significant wave-of yellow light relative to that of green length effect. Blue light measurements inlight after day 6 (Fig. 4) is an interest- general produced smaller relative indicesing phenomenon. Although the relative in- of refraction than did the green and yellowcrease is quite small and could be due to light, as indicated by Fig. 4. Theoreticallyexperimental error alone, it is also possible there should be no wavelength effect sincethat a change in the relative concentra- wavelength variations have already beentions of various pigments occurred after accounted for in the parameter p.Butday 6.Yentsch and Vaccaro (1958) have the assumption has been made that Ks measured sharp decreases in the chloro- Qext,which is true only for nonabsorbingphyll : carotenoid ratioinalgalcultures spheres.Sincecellsof1. galbana areduring nitrogen deficiency. Such a varia- slightly greenish yellow and contain nuclei,tion in the pigmentation ratio could ex- this assumption is less valid for bluewave-plain the relative decrease in the apparent lengths and other absorption bands thanindex of refraction using green light, since for green and yellow. ,3-caroteneismore absorptive of green Additionalquantitativeanalysesarewavelengths than of yellow.Since nutri- needed to define the role plant pigmentsents were not measured during the experi- play in the selective absorption of light.ment, this hypothesis is pure speculation, Absportion spectra for plant pigmentsarebut through techniques similar to those usually measured ina solvent such asdescribed above, the sensitivity of the ap- ether or acetone (Strickland 1965). Theparent index of refraction to changes in wavelengths of the absorption maxima ofpigment ratios may prove to be useful in various chlorophylls increase when mea-monitoring the effective nitrogen levels in suredin vivoover measurements in ethergrowing algal cultures. solutions (French 1960). In this experiment, most of the spectral In vivoabsorption measurements are atvariation in the refractive ratio fell within best difficult due to the overlapping ofthe limits of the 2% relative experimental absorption bands of several pigments anderror.This means that either the spectral to selective scattering. Anin vivospectralfluctuations are due to experimental error absorption curve for a suspension ofChlo-or that another factor (e.g. nutrient defi- rella,a phytoplankter that may have simi-ciency) needs to be considered; further lar spectral characteristics to those of I.experimentation is required to resolve this galbana,showed a relative absorption coefficient of blue light (4,360 A) to be aboutquestion. The total experimental error can 13 times that of green light (5,460 A) andonly be estimated because of the difficulty about 10 times that of yellow light (5,780involved in assessing the errors induced A) (Latimer and Rabinowitch 1956). Theby the simplifying assumptions. Pak (1970) green and yellow wavelengths fell nearand Carder (1970) have analyzed the cali- the minimum of the visible portion of thebration and measurement errors involved spectral absorption curve, so littleerrorin similar applications of the Brice-Phoe- is introduced by assuming thatIsochrysisnix and Coulter counter and found them cellsare nonabsorbing for these wave-to be about 3 and 5% respectively.It is lengths.Increases in spectral absorptionestimated that the total experimental error result in decreases in the apparent indexin values of (m-1) will fall in the range 162

838 CARDER, TOMLINSON, AND BEARDSLEY

of 8 and 13%, and values of (m -1) will REFERENCES probably be slightly on the low side. BEARDSLEY, G. F., JR. 1966.The polarization The real (nonabsorptive) part of the of the near asymptotic light field in seawater.Ph.D. thesis, Mass. Inst. Technol., Cam- index of refraction ofI. galbanarelative bridge. 119 p. toseawater(S = 35%,,)variedsignifi- H.PAK,K. CARDER, AND B. LUNDGBEN. cantly with the mean particle diameter 1970. Light scattering and suspended par- which changed with time by as much as ticles.J. Geophys. Res. 75: 2837-2845. 9%. This was probably due to concurrentBRYANT, F. D., B. A. SEIBER, AND P. LATIMER. 1969.Absolute optical cross-sections of cells changes in the surface area: volume ratio and chloroplasts.Arch. Biochem. Biophys. since the cell wall represents one of the 135: 79-108. most refractive parts of the cell.If the BURT, W. F.1956. A light scattering diagram. assumption that algal absorption of yellow J. Mar.Res.15: 76-80. CARDER, K. L. 1970.Particles in the eastern and green light is negligible were incor- equatorial Pacific. Ocean:Their distribution rect,it would cause these experimental and effect upon optical parameters.Ph.D. values of refractive index to be slightly thesis,Oregon StateUniversity,Corvallis. low.The literature,however, indicates 146 p. C. F. BEARDSLEY, JR., AND H. PAK.1971. that absorption of these wavelengths by Particle size distributions in the eastern equa- the primary algal pigments is small. The torial Pacific.J. Geophys. Res. 76: 5070- indicesofrefraction found using blue 5077. light (4,360 A) could not be used due to DAVIS, H. C., AND R. UKELES. 1961. Mass its high absorption by the algal pigments culture of phytoplankton as foods for meta- which caused an apparent lowering of the zoans.Science 134: 562-564. DEIRMENDJIAN, D. 1963. Scattering and polar- refractive index. ization propertiesof polydispersed suspen- Although thein vivomeasurement of the sions with partial absorption, p. 171-189.In refractive index of one species of marine M. Kerker [ed.], Electromagnetic scattering, phytoplankton does not allow many gen- v. 5. Pergamon. FRENCH, C. S.1960.The chlorophyllsin vivo eralizations concerning the light-scattering andin vitro, p.252-297.In W. Ruhland propertiesof suspended marinepartic- [ed.], Encyclopedia of plant physiology, v. 5. ulates,it does suggest a few.SinceI. Springer. galbanais an unarmored type of phyto- JERLOV, N. G. 1953.Particle distributionin plankter, its index of refraction is near the the ocean.Rep. Swed. Deep-Sea Exped. 3: 73-97. lower limit of the range of indices expected . 1988. Optical oceanography.Elsevier. for marine phytoplankters and representa- 194 p. tive of those indices expected for soft or- LATIMER, P., AND E.I.RABINOWITCH. 1956. ganisms.Since the light scattering for a Selected scattering of light by pigments-containing plantcells.J. Chem.Phys. 24: 480. given geometrical cross-section of marine MIE, G.1908.Beitrage zur optik truber Medien, particulates increases with depth (Pak et speziellKolloidalenMetal-losungen. Ann. al.1970), this suggests that the harder, Phys. 25: 377. OCHAKovsxY, Y. 1966. On the dependence of more refractive parts of organisms are all the total attenuation coefficient upon suspen- that remain of them at depth.Conse- sionsin the sea.U.S. Dep. Commerce, J. quently, the mean refractive index relative Publ. Res. Ser., Rep. 36816, Mon. Cat. 13534. PAK, H.1970.The Columbia River as a source to seawater of suspended marine particu- of marine light scattering particles.Ph.D. lates is expected to be greater than 1.036. thesis,OregonStateUniversity,Corvallis. Optical measurements of diatoms, coc- 110 p. C. F. BEARDSLEY, JR., G. R. HEATH, AND colithophores,etc. need to be made to H. CURL. 1970.Light-scattering vectors of better understand the range ofeffects some marine particles.Limnol. Oceanogr. that various phytoplankters have on the 15: 683-687. espe- PARKE, M. 1949.Studies on marine flagellates. opticalpropertiesof the oceans, J. Mar. Biol. Ass. U.K.28: 255-286. cially in regions of high standing stock of SCHLEMMER, F. C. II, AND K. L. CARDER.1972. phytoplankton. Particles as indicators of circulation in the 16:s

REFRACTION INDICES OF MARINE PHYTOPLANKTERS 839

eastern Gulf ofMexico.Trans.Amer. Geo- STRICKLAND, J. D. H. 1965. Production of or- phys. Union53: 424. ganic matter in the primary stages of the SHELDON,R. W., AND T. R. PARSONS. 1967a. marine foodchain, p.477-610. In J.P. A continuous size spectrum for particulate Riley andG. Skirrow[eds.], Chemical ocean- matter in thesea. J.FishRes. Bd. Can. ography, v.1.Academic. 24: 909-915. VAN DE HULST, H. C.1946.Optics of spherical , AND . 1967b.A practicalmanual on the use of the Coulter counter in marine particles.Duwaer and Zonen, Amsterdam. research.Coulter electronics, Toronto. 66 p. 87 p. SPILHAUS, A. F., JR.1965.Observations of light YENTSCH, C. S., AND R. F. VACCARO.1958.Phy- scattering in sea-water.Ph.D. thesis, Mass. toplankton nitrogen in the oceans.Limnol. Inst.Technol., Cambridge. 242 p. Occanogr. 3: 443-448. 164

RLO-2227-T12-21

Pacific Science(1973), Vol. 27, No.2, p. 156167 Printed in Great Britain

Growth of a Sea Urchin, Allocentrotus fragilis, off the Oregon Coast,

JAMES L. SUMICH2 AND JAMES E. MCCAULEY3

ABSTRACT: Allocentrotus fragilis (Jackson) was obtained from six stations at depths of 100 to 1,260 m on the continental shelf and upper slope off Newport, Oregon. Ages and growth rates ofA. fragiliswere determined by two methods: (1) from size-frequency distributions of trawl collections from 200 m, and (2) from growth zones onskeletal test plates. Collections from other depths were not adequate for size-frequency analyses. Gonad indices ofA. fragilisfrom 200 mwere used to determinespawning periodicity and frequency.A semiannualfrequencywas suggested, with spawning occurring in early springand earlyautumn. No individuals collected below 400 m were reproductively mature. A procedure was developed to make growth zones ofthe skeletal test platesvisible. Dark growthzones are thought to correspondto semiannualperiods of growth, one-half the number of dark growthzones indicatingthe urchin's age. The growth curve ofA. frugilisfrom 200 m, which was plotted from the mean testdiameter of age groups defined by test plate growth zones, shows a good leastsquares fit to von Bertalanffy's growth equation. Growth rates determined from plate growth zones appearedto be similarforA. fragilisfrom 100 to 600 m, but decreased for specimens from 800 to 1,260 m. The asymptoticsizedecreased with increasing depth below 200 m.

THE PINK SEA URCHINAllocentrotus fragilis Ages and growth rates of sea urchins have (Jackson) inhabits the continental shelf and been determined by one or a combination of the upper slope of the west coast of North Americafollowingmethods:(1) tagging and observa- from Baja California, to Vancouver Island,tion of selected individuals in their natural British Columbia, from a depth of 50 to 1,150habitat, (2) laboratory observation of urchins, m (Mortensen, 1943). The physiology, repro- (3) size-frequencydistributions,or (4) analyses duction, and ecology of this species have beenof cyclic growth indicators in spines and test studied by A. R. Moore (1959), Boolootian plates. et al. (1959), and Giese (1961); and McCauley Individual tagging, such as that used by and Carey (1967) have discussed and summari-Ebert(1965),is an effective way of following zed its distribution.Allocentrotus frargilisoccursthe growth of individual urchins, but is restric- in large numbers off Oregon where it is foundted to areas where urchins may be collected, on unconsolidated sediments, ranging from tagged, and replaced for later recovery. Labora- sand toclaysilts (McCauley and Carey, 1967); tory growth studies are normally restricted to and it is the dominant benthic organism, in shallowwaterurchins,because deepwater terms of biomass, at 200 in off Newport, species suchasA. fra,gili.r often do not survive Oregon (Carey, 1972). Yet, no information onthe rigors of collection or do not survive long size specific growth and longevity ofA. fragilis in the laboratory (Boolootian et al.,1959).tiles or other deepwater echinoids is available. (personalcommunication)kepta few A. fragilis from 200 m off Oregon in aquaria for I Researchsupported in part by the U.S. Atomic about 5 months but made no conclusions about Energy Commissionunder contractAT (45-) 2227 Task growthrates.Size-frequencydistributions, Agreement12.Itisassigned manuscript number RLO 2227-T12-21. Manuscript received 8 August 1972. used successfullyby Fuji(1967) and Ebert 2 Grossmont College, El Cajon, California 92020. (1968)to establish the ages and growth rates 3Oregon StateUniversity, Corvallis, Oregon 97331. of two intertidal species of Strongylocentrotus, 156 1.6 3

Growth ofa Sea Urchin-SUMICH ANDMCCAULEY 157 require periodic large samples representative of a population. METHODS AND MATERIALS A variety of patterns observed in the hard Allocentrotus fragilis was collected by the parts of various marine invertebrates has beenOregon State University Department of Ocean- related to periodic variations in the structure or ography as part of regular sampling programs. pigmentation during growth. These are exem- Samples from depths of 100, 200, 400, 600, 800, plified by growth lines of pelecypod shellsand 1,260 m were obtained along a line extend- (Clark, 1968; House and Farrow, 1968) anding west fromNewport,Oregon, near lat seasonal pigment bands in the shells of abalone44°39' N, with otter and beamtrawls.Popula- (Olsen, 1968). Many echinoids also exhibittion density at 200 m was estimated from an alternating light and dark bands in the plates ofodometer-wheeled beam trawl collection. the test and in spines which are superficially Theurchins,preserved in neutral formalin, similar to the annual growth rings found in were returned to thelaboratory,scrubbed with woody plants. a stiff brush to remove thespines,and measured Some echinoids, includingA. fragilis,havefrom ambulacrum to the opposite interambu- spines which are too small for the ring-count lacrum. The two halves of an emptied test were method. The dark bands observed in the testrinsed, then placed in 25-percent household plates of echinoids have received less study asbleach (e.g., Clorox) formaceration,but were age indicators than have spine rings, yet inter-removed from the bleach solution before pretation is theoretically much simpler. Break-disarticulation of the test plates occurred. One age and loss of plates do not normally occur, interambulacralcolumn andtheadjacent whereas spine loss and damage is common ambulacral column of test plates extending from (Ebert,1967;Weber,1969). The earliestthe peristome to the periproct were separated formed plates are always associated with thefrom the remainder of a test half. The plates urchin for its entire nonlarval life and, therefore,were arranged in order, flat, on a 1-by-3-inch should represent the age of the urchin moremicroscope slide with the inner side of the test accurately. plates up, and heated on a hot plate at 85° C for Deutler (1926) ground the plates of a variety3 hours to remove all moisture. After cooling, of echinoids to make the dark bands visible,the plates were wetted withxylene,then satura- and H. B. Moore (1935), using a similar method ted withPermount,and allowed to dry. Dark with the genital plates ofEchhinus esculentus bands were counted under a 6 x dissecting Linnaeus, found good agreement between the microscope. number of dark bands and age of the urchins as determined from size-frequency distributions. A darkly pigmented zone was added to each RESULTS plateduringthe summer, and alighter At the six stationssampled,A. fragiliswas nonpigmented zone was added during thetaken consistently only at the 200-m station winter. (Table 1). The values are given only as a rough A much less tedious method of making darkindicationofA. fragilis abundance at each bands visible,described by Jensen (1969),station,because trawling times were not always involves clearing the plates with an organic the same. solvent. She found a pattern of test-plate-band A beam trawl collection made on 15 March formation for the North Atlantic urchinStrongy- 1970 at the 200-m station yielded 687 A. locentrotus droebachiensis (D.F. Miiller) that sug-fragilis. Approximately 3,200 m2 of bottom area gested a maximum number of dark zones inwere sampled. Only otter trawl samples were the test plates representing the urchin's age inobtained at the other stations. McCauley and years. Carey(1967) haveshown,using a Benthos For the present study, a modification ofTime-DepthRecorder,that the otter trawl Jensen's method was used in conjunction with may be off the bottom as much as 75 percent of size-frequency distribution analyses to deter-the time, making estimates of the area sampled mine the age and growth rate ofA. fragilis. extremely difficult andunreliable.Even so, data 166

158 PACIFIC SCIENCE, Volume 27, April 1973

Table 1 SUMMARY OF Illseenlrsttus fragths CoLLI CTIONS AT LAT4439' NSINCE APRIL 1963

STATION DEPTH 100 to 200 m 400 m 600 in 800 m 1,260 m

Number of Otter Trawls 10 38 4 6 26 6 Trawls with A. fragili.r 1 38 3 3 4 1 Percent with A. fragilis 10 100 75 40 15 17 %laximum Number _9. fragilisfTrawl 16 750 35 93 89 5

10 Z 0LL it30- I w SEP. 1969 nJ in201- I

z 10 11 ,

0 n I 0 10 20 30 40 50 60 70 80 TEST DIAMETER (mm)

FIC. 1. Time series of_lllocentrotus fragilissize-frequency distributions from 200 to. Dashed lines intersect the means of well-defined size classes which indicate assumed size increase with time.

provided in Table 1 show that A. fragilis were much more abundant at the 200-m station than Size-Frequency Distribution Analysis at the otherstations.Although trawling times The usefulness and accuracy of size-frequency were notconsistent,shallow stations (100 and distributions as indicators of age depend on the

200m)tended to be trawled for shorter times,following assumptions:(1) size-frequency peaks so that estimates of abundance at these stations of successive age groups must be discernible; are conservative. (2) growth rates within an age class are normally The smallest A. fragilis test diameter, 8mm, distributed;(3) the population sample should was assumed to be the minimum size catch capa- include representative numbers from all size bility of either the beam or the otter trawl nets, classes present in thepopulation;and (4) and test diameters smaller than the stretched spawning must occur periodically witha size of the net liner (12 mm) were probably spawning duration short in relation to the length not representative of the sampled population. of the reproductivecycle. Trawlcollections of FiveA. fragilis from 1,260 m (lat 44°36' N, A. fragilis from stations other than 200 m were long 125°02' W) extend the bathymetric rangeneither numerous not periodic. Size-frequency 110 m deeper than 1,150 m (reported by Mor- analyses of these collections were not attempted tensen,1943). due to small numbers of individuals collected 16-11

Growth of a Sea Urchin-Suniici-t AND MCCAULEY 159

E 064.00

20 40 60 80 TEST DIAMETER (mm) Fic. 2. Variation of gonad wet weight with test diameter of Allocenirotus fragilisfrom 200 m.

per trawl and to the lack of successive periodic size increase was approximately twice as large as collections at any one station. that computed from the mean size difference The test diametersofA. fragilisfrom 11 between successive age classes (10 to 11 mm). trawl collections taken between October 1968The lack of agreement between the results of and March 1970 from the 200-m station were the two methods suggested that the growth rate measured.The size-frequency distributions forcalculated for the 2-month interval from Sep- the September andNovember,1969, samples tember to November, 1969, was substantially represent complete trawl collections and aregreater than the average annual growth rate, shown inFig.1. The remaining trawl collec-that the size classes determined from size- tions were either subsampled at sea or consistedfrequency analysis corresponded to semiannual only of large adult urchins exhibiting no dis- rather than annual recruitment classes, or that cernible size classes. the trawls sampled separate populations. The trawl collections taken in September and Very little information exists about monthly November, 1969, have two well-defined sizeor seasonal growth rate fluctuationsof echi- classes of small urchins, each with means cen- noids. Fuji (1967) obtained monthly measure- tered 10 to 11 mm apart. By assuming that each ments of the diameters of five-year classes of the peak represented a separate age class and that intertidal Strongylocentrotus intermedius (A. Agas- the two November peaks were derived from siz) and found that each year class exhibited a theslightlysmallerSeptember peaks, weconstant annual growth rate which differed determined two different measures of A. fragilis from the other year classes. Similar observations growth rate. If the increase in the mean testof deepwater echinoids have not been made. diameter of each size class during the 2-month interval between the September and November Reproductive Periodicio, collections was 3.2 mm for the smaller peak and 2.6 mm for the larger peak, this was equi- Reproductive periodicity was studied only valent to approximately 20 mm of annual test in specimens from the 200-m station, as the diameter increase for the small size class andremaining stations had not been sampled 16 mm for the largerpeak.This assumed annual successively. 168

160 PACIFIC SCIENCE, Volume 27, April 1973

Table 2

SUMMARYOFAllocentrotus fragilisGONAD WETWEIGHTSFROM+STATIONS BELOW 200 METERS

DEPTH NUMBER OF GONAD WET WEIGHT (g) (m) SPECIMENS MEAN-L-1 SD.RANGE

400 14 0.22_+ 0.43 0.00-1.73 600 21 0.02-+0.050.00-0.22 800 21 0.02 -10.040.00-0.14 1,260 5 0.00+0.03 0.00-0.01

(15) 3.00r (I0)(7) X Q 2.00 z

1.00r- 0 -I-- 0 I I I I I A S 0 N F M A 1969 1970 FIG. 3. Monthly gonadindex variationofAlocentrolus fragilis greater than 49 mm in test diameter. Range of values is shown as a central verticalline.One standard deviation on each side of the mean is represented by a rectangle. The numbers of specimens are shown in parentheses.

The minimum size at sexual maturity was animal(Boolootian, 1966). Accuratewet estimated by plotting gonad wet weight as aweights quite often could not be obtained from function of test diameter for 200-m stationA. fragilis because of crushing and abrasion of specimens (Fig.2). The maximum gonadal the tests and spines during trawling and because weight was low for animals less than 45-mm the amount of fluid contained within the test or test diameter, then increased rapidly for speci- adhering to it could not becontrolled. There- mens larger than 50-mm diameter. The actualfore, test diameter rather than test volume or minimum size at sexual maturity may be less total wet weight was selected as a simple, but than 50 mm, but an unexplained paucity ofeffective,indicator of size because the urchins specimens with test diameters between 35 and used were approximately the samesize.A gonad 50 mm restricted the study of gonadal develop- index was computed from the following for- ment in this range. mula: Gonad indices, determined from the gonadal gonad wet weight gonad index = x 100. biomass, have been widely used as indicators of test diameter reproductive state and spawning period of echinoids (Boolootian et al., 1959; Giese, 1961; The gonad index values ranged from less than Boolootian, 1966; Fuji, 1967). The gonad index 0.1 for immature specimens to 3.3 for mature has generally been calculated as the ratio ofindividuals. gonad wet weight to the total wet weight of the Gonad wet weight determinations suggested 169

Growth of a Sea Urchin-SuMicx AND MCCAULEY 161

that some urchins from the 400-m station wereexhibited in any one plate of an interambulacral reproductively mature but all specimens fromcolumn. Plates near the peristome are the oldest. 600, 800, and 1,260 m had gonad indices of less As the urchin grows, its existing plates increase than 0.3 and appeared immature (Table 2).in width and new plates are formed in the apical Although data for the deep stations are limited,region. As more plates are formed-they migrate gonad indices suggest thatA. fragilisdoes notfrom the apical region toward the ambitus, achieve sexual maturity below 400 m. where they increase greatly in width but not in Specimens from 200 m which were largerheight. It is in the wide lateral areas of the plates than the estimated minimum size at sexualthat growth zones are best seen and counted. maturity show marked semiannual peaks inThis growth process is shown in Fig. 4. gonadal weight, which strongly suggests two As the test diameter continues to increase, spawning seasons per year: one in late sum-older plates migrate past the wide ambitus mer and one in later winter (Fig. 3). Two ageregion toward the more narrow peristome. classes are thus spawned each year. EarlierLateral growth ceases as they pass the ambitus studies on the reproductive periodicity of A.because the test area in which they are now fragilisat Monterey Bay, California, establishedlocated is no longer increasing in size. Cessation a spawning period between January and Marchof lateral growth halts the formation of growth (Boolootian et al., 1959; Giese, 1961; Booloo-zones in these plates; but zones continue to be tian, 1966), but no reference vas made to anformed in the plates above the ambitus. Thus, autumn spawning period. There the spawning the oldest plates near the peristome exhibit only of -I. fragilisis closely related to the onset ofearlier formed growth zones, not more recent localupwelling, a condition common forzones. Younger plates, near the apical area, do benthic invertebrates which produce planktonicnot exhibit early growth zones but show only larvae. The spawning periods ofA. fragilisthe more recent, some of which are not exhib- off Oregon also appear to be closely associatedited by older plates. with periods of high primary productivity of the We counted growth zones by beginning with surface waters (Anderson, 1964). the center of the oldest plates near the peristome, A. R. Moore (1959) found that at equivalentcounting growth zones laterally, moving up the temperaturesA. fragilisandS. franciscanus corresponding growth zone to a younger plate, (A. Agassiz) larvae develop similarly up to atand counting laterally again. This process was least the 5-day stage. After 5 days, his experi-repeated until all growth zones within an inter- ments were terminated. Details of the laterambulacral series had been counted. (See Fig. 4.) larval development ofA. fragilisare not known, The maximum number of dark growth butS. franciscanuslarvae required 62 days tozones was determined for one interambulacral complete development from fertilization toplate series of each urchin examined. The least metamorphosis (Johnson, 1930). Newly meta-number of dark growth zones found was one, morphosedS. franciscanurare less than 1 mmin specimens 8 to 9 mm in diameter; the in diameter. If the rates of late larval develop-greatest number was 15 in a specimen 80 mm in ment of both species remain nearly equal, larvaldiameter. Both specimens were collected from metamorphosis andsettlingof A.fragilisthe 200-m station. The number of dark growth would be expected approximately 2 monthszones of the individuals was related to the after spawning. Off Oregon, settling woulddistribution of test diameters. Some scatter and probably occur in April-May and October-overlap occurred in specimens larger than November of each year. 50 mm, but the size-frequency peaks less than 40 mm in diameter generally were composed of specimens with the same number of growth Growth Zone Counts zones; specimens of successive age classes had Visible dark bands, or growth zones, in testconsecutive numbers of dark growth zones. plates of 5 to 30 specimens from each trawl The test diameters were plotted as a function collection were counted. The maximum num-of the maximum number of dark zones of the ber of growth zones inA. fragilisis nottest plates of each specimen. A growth curve for II 11 PS 27 170

162 PACIFIC SCIENCE, Volume 27, April 1973

APEX

I'f)l AMBITUS 5_ (qrT

((( 4W 0 OW

11o)

PERISTOME

A A

X 1.7

B B' FIG. 4. Photographs and tracings of interambulacral columns of two Allocentrotus fragilis specimens from 200 m: A, from a specimen 53 mm in test diameter; B, from a specimen 17 mm in test diameter. The dark growth zones are numbered in A' and B' according to the order in which they were formed. 171

Growth of a Sea Urchin--Sun1Icx AND MCCAULEY 163 80

At = 113.56 [-e-1.54(age -.15) Variance = 506

0 No. Growth Zones I 2 4 6 8 10 12 14 Age (years) I 2 3 4 5 6 7 No. Examined 3 6237 I I I II 9 II 9 56 2 1 Fu;. 5. Lcast-squaresfit ofvon Bertalanffy growth curveforAllocentrotus frayilisfrom200 m. Means 1!1 standard deviationof eachgrowth zone class are shown.

A. fragilis at 200 m was constructed using the these growth rates do not agree with the growth test diameter means and standard deviations of rates estimated from the mean annual size each "growth zone class" (Fig.5).The von increase of the age class. Bertalanffy type of growth curve was fitted to the mean growth zone sizes using the method of Synopsis of Growth Rate Determinations leastsquares described by Tomlinson and Abramson(1961).Urchins from 200 m have Agreement between the results of gonadal approximate growth rates of 9 mm per growth studies and of the different methods of growth zone period for 15-mmurchins,and 7 mm perrate determination can occur only if each age growth zone period for 30-mm urchins. class and each new dark growth zone is added These growth rates compare favorably with semiannually.H. B. Moore(1935)considered those determined by the mean size difference of the growth zones of Echinus esculentus to be bands successive ageclasses.This means that theoffood-derived pigment which were incorpora- frequency of spawning and the frequency ofted seasonally into growing skeletal material growthzone formation are thesame.However, during the season of rapid intertidalplant rr-s 172

164 PACIFIC SCIENCE, Volume 27, April 1973

Table 3

FREQUENCY DISTRIBUTION OF JEST DIAMETERSOFAll cenIroIus fragil s BELOW 200 METERS

TEST DIAMETER 400 m 600- m 600,m 800 m 1,260- m (mm} JUNE1966 APRIL 1963 JUNE1964 AUGUST 1966 JUNE 19166

16 17 III 18 19 11I 1111 20 l II1 21 l III 22 11II 23 24 -- - 1 25 2 26 1IV 1111 2V 27 6 I 3IV 1V 28 -- 4 2111 21V 29 11V 11IV 1 1 30 11V 81V 1 2 31 2 8 8V 1 VI 32 - 6 2IV 7V 33 8 21V 4VI 34 -- 6 1 3V 35 1V 8V 2 6V

36 2V 3 3 - 9VI 37 1V 4 3V 6IV 38 1 3V 61V 9VI 39 - 4V 2 3V1 2VIII 40 1 2 1V 3 41 1 3 1 3V11 42 3 IV 5 43 2 - 1V 2 44 3 45 1 VI I 2VIII 46 2 VI 47 3 lix 48 --- 2VII 49 1 50 - 51 2 52 2V11 53 3VII 54 55 56 57 iix 58

NOTE: Arabic numerals represent numbers of animals for each size class. Roman numerals represent numbers of dark growthzones found in urchins of that size class.

growth. Growth zones in the plates of A. ExaminationofA. fragilis gut contents shows fragilis seem to reflect semiannual fluctuations that this species is a particulate detrital feeder of the amount of pigmented organic materialand could incorporate into its test seasonally available from surface productivity, which fluctuating amounts of plant pigment produced suggests that the two pigmented zones formed during spring and fall phytoplankton blooms. annually are the result of semiannual plankton However, nopublishedinformation on the blooms in surface water. chemical composition of the pigmented bands 173

Growth of a Sea Urchin-SuMICH AND MCCAULEY 165

80r_

4 6 8 10 12 14 NUMBER OF GROWTH ZONES Fu;. 6. Compilation of /Illocentrotus 1i-agi/isgrowth curves from all stations. The 200-m growthcurve is from Fig. 5. Other curves are fitted byeye to data presented in Table 3. of any echinoid test plates is availableto con-other strongylocentrotoid urchins issparse and firm this. to a large extent based on circumstantial evi- If two growth zonesare formed each year,dence, some comparisons may be made. Fuji the data in Fig. 5 suggest growth rates of 20 mm(1967) followed the growth of S.intermediusfor per year for 20-mm urchins and 18mm per5 years, until the urchins were near the maxi- year for 30-mm urchins from the 200-m station.mum size for that locality. Swan (1961) found, These growth rateswere essentially equal tousing size-frequency studies and laboratory those calculated from the shift of themean sizegrowth studies, that S.droebachiensisattained a of each age class between Septemberand test diameter of 52 to 54 mm in 4 years, but November, 1969. Jensen (1969) reported 11 annual growthzones Although information on maximum ages forin the test plates of a 51-mm S.droebachiensis 174

166 PACIFIC SCIENCE, Volume 27, April 1973

specimen from Norway.Strongylocentrotus pur- LITERATURE CITED puratusare thought to live at least 10 years (Ebert, 1967). ANDERSON, G. C. 1964. The seasonal and geo- The maximumA. fragilisage of 7.5 years, graphic distribution of primary productivity determined from growth zone counts of an off the Washington and Oregon coasts. 80-mm specimen, compares favorably with the Limnol. Oceanogr. 9: 284-302. reported maximum ages of other strongylo-BOOLOOTIAN, R. A. 1966. Reproductive physio- centrotoids. Individuals up to 88 mm in test logy, p. 561-614.In:R. A. Boolootian [ed.] diameter were collected in this study and larger Physiology of Echinodermata. John Wiley specimens have been reported. Thus, this and Sons, Interscience, New York. xviii + species may be expected to live at least 10 years. 822 p. Age and growth ratesfor A. fragilisfrom theBOOLOOTIAN, R. A., A. C. GIESE, J. S. TUCKER, remaining stations were computed from data and A. FARMANFARMAIAN. 1959. A contri- presented in Table 3 in a manner similar to those bution to the biology of a deep sea echinoid, at 200 m, but data were insufficient to apply the Allocentrotus fragilis(Jackson). Biol. Bull., Tomlinson and Abramson (1961) least squares Woods Hole 116: 362-372. method of curve fitting. Growth curves forCAREY, A. G., JR. 1972. Ecological observa- these stations were fitted by eye. These curves tions on the benthic invertebrates from the and the fitted curve of the 200-m station are central Oregon continental shelf, p. 422-443. shown in Fig. 6. The age range ofA. fragilis In:A. T. Pruter and D. L. Alverson [ed.] The from the 100-m station was small and no growth ColumbiaRiverestuaryandadjacent curve was plotted; however, its growth rate oceanwaters :bioenvironmentalstudies. appears to be similar to that at 200 m. The Univ. Wash. Press, Seattle. xiii + 868 p. growth curves for all stations were assumed toCLARK, G. R., II. 1968. Mollusk shell: daily have the same x-intercept, representing the growth lines. Science 161: 800-802. assumed duration of the larval stage. DEUTLER, F. 1926. Ober das Wachstum des Little apparent difference exists between the Seeigelskeletts. Zool. Jb. 48: 119-200. growth rates forA. fragilisfrom 100, 200, 400,EBERT, T. A. 1965. A technique for the indivi- and 600 m. This similarity suggests that food dual marking of sea urchins. Ecology 46: availability does not vary significantly among 193-194. these stations. However, the growth rates for 1967. Negative growth and longevity the urchins from 800 and 1,260 m are markedly in the purple urchinStrongylocentrotus pur- less than the rates for the more shallow stations. puratus(Stimpson). Science 157: 557-558. Maximum age and maximum test size are also -. 1968. Growth ratesof the sea urchin less at the deeper stations. It is tempting to Strongylocentrotus purpuratusrelated to food propose such factors as low food availability or availability and spine abrasion. Ecology 49: poor larval settlement as possible causes for the 1075-1091. reduced growth rates, smaller maximum sizeFuji, A. 1967. Ecological studies on the growth and age, and reduced densities exhibited by and food consumption of Japanese common A. fragilisat the deeper portions of its bathy- littoral sea urchin,Strongylocentrotusinter- metric range. However, the general lack of medius (A.Agassiz). Mem. Fac.Fish., information regarding the feeding habits, sub- Hokkaido 15: 83-160. strate preferences, patterns of larval settlement,GIESE, A. C. 1961. Further studies onAllocen- and responses to hydrologic conditions of A. trotusfragilis,a deep-sea echinoid. Biol. Bull., fragilisand many other deepwater species of Woods Hole 121: 141-150. benthic animals makes such suggestions pre-HousE, M. R., and G. E. FARROW. 1968. Daily mature and emphasizes the need for further growth banding in the shell of the cockle, research in these areas. Cardiun edule.Nature 219: 1384-1386. 17>

Growth of a Sea Urchin-SUMICH AND MCCAULEY 167

JENSEN, M. 1969. Age determination of echi- dae, and Echinometridae, p. 183-193, 254. noids. Sarsia 37: 41-44. C. A. Reitzel Co., Copenhagen. JOHNSON, M. W. 1930. Notes on the larvalOLSEN, D. 1968. Banding patterns of1laliotis development ofVtrong9,Iocentrotus franciscanus. rufescensas indicators of botanical and animal Pub]. Pugct Sd. Mar. (Biol.) Sta. 7: 401-411. succession. Biol. Bull., Woods Hole 134: MCCAULEY, J. L, and A. G. CAREY, JR. 1967. 139-147. Echinoidca of Oregon. J. Fish. Res. Bd.SWAN, E. F. 1961. Some observations on the Can. 24: 1385--1401. growth rate of sea urchins in the genus MOORE, A. R. 1959. On the embryonic develop- Strongylocentrotus.Biol. Bull., Woods Hole ment of the sea urchinAllocentrotu. fragilis. 120: 420-427. Biol. Bull., Woods Hole 117: 492-496. TOMLINSON, P. K., and N. J. ABRAMSON. 1961. MOORE, H. B. 1935. A comparison of the Fitting a von Bertalanffy growth curve by biology ofEchinus esculentusindifferent least squares. Fish. Bull. Calif. Dep. Fish habitats. Part II.J. Mar. Biol. Ass. U.K. Game 116. 21: 109-128. WEBER, J. N. 1969. Origin of concentric band- MORTENSEN, T. 1943. A monograph of the ing in the spines of the tropical echinoid Echinoidea. Vol. 3, Pt. 3. Camarodonta II, I I eterocentrotus.Pacif. Sci. 23: 452-466. Echinidae, Strongylocentrotidae, Parasalenii- 176

EstimatingPrecision for the Method of Standard Additions

ingvar L. Larsen,' Norbert A. Hartmann,2 and Jerome J. Wagnerl Oregon State University, Corvallis, Ore. 97331

An estimate of the uncertainty term expected in the is determined by extrapolating a line to theabscissa. (If method ofstandard additions using linear regression the line is not linear, often a transformation can be per- analysis is presented. The method agrees favorably with formed such as the conversion of light absorption to ab- the standard deviation for values which are not corrected sorbance.) The best line which can be fitted to the data for a blank as well as with the population standard error minimizes the sum of the squares of the vertical distance of difference for corrected samples. Analysis for zinc in between data points and the constructed line, referred to an environmental sample yielded a concentration range as the line ofleast squares (4). within the expected value. The least squares linecan be describedby the equation k = MX + I (1) where' is the value predicted (absorbance) by the equa- The precision for an experimental process is usually de- tion for a given value ofX(concentration),Mis the slope terminedthrough replication. When applied to calibration of the line, and I is the intercept of the line with the ordi- linedata, however, sample replication may not only be in- nateaxis. Using data obtained from an analytical deter- sufficientbut even misleading (1). In addition, when ap- minationof treated and untreatedsamples,the slope of plied to the method of standard additions, replication be- the least squares line can becalculated as follows: comes tedious as well as time consuming.An estimated nEXY - EXEY error termbased upon linear least squares regression ap- Slope, M - (2) pliedto the method of standard additions is presented as nZX2 -(2;X)' an approximationto the experimental error. The method of standard additions is often used to de- where X refers to the concentration of the standard solu- and is is termine theconcentration of an element in an interfering tion, Y isthe linear instrument responsereading, matrix(2). Caution, however, should be exercised when the total number of readings made. using thistechnique to ensure that the initial calibration Equation 1 is referred to as the regressionofYonX, X of the instrumentbe accomplished with standards con- being the independent variable andY the dependent vari- tained in as similara matrix as the sample, both chemi- able. The X values are assumed to be known without cally and physically, in order toestimate the appropriate error,whereas the Y values are randomly distributed "blank" correction. In this way correction for light scat- about some mean Yvalue(5, 6). tering due to the matrix canbeapproximated.Rains (3) The intercept of the line with the ordinate axis is calcu- discusses some aspectsof elimination or controlof inter- lated as follows: ferences in atomicabsorption spectrophotometry. Intercept, I = Y - MX (3) In themethod of standardadditions, a small known concentrationof the desired element is increasingly added where ? is the arithmetic mean of thetotal Yreadings, X to several samples ofthe unknown testsolution, and the is the arithmetic mean of thetotal Xvalues, and M is the resultingsolutionsas well as anuntreatedone are then slope of the line. analyzed. The response readings from the particular in- An estimate of the variability of the data points about strumentare then plotted linearlyagainstthe added con- the least squares line is the standard error of regression centrationsand the amount of unknownelementpresent and is calculated as follows:

I School of Oceanography. 2 Department of Statistics. (4) C. L. Grant, Statistics helps evaluate analytical methods. "Develop- ments In Applied Spectroscopy," Vol. 6. Proc. of the 18th annual mid-American spectroscopysymposium, Chicago.15-18 May (1) F. J. Linnlg and John Mandel, Anal. Chem., 36(13). 25A (19641. 1967, W. K. Baer, A. J. Perkins, and E. L. Grove, Ed., Plenum (2) G. Manic and T. C. Rains, in "Analytical Flame Spectroscopy." Press, New York, N. Y., 1968, pp 115-26. R. Mavrodineanu, Ed., Philips Technical Library, Eindhoven, The (5) John Mandel and Frederic J.Linnig, Anal. Chem, 29, 743-749 Netherlands, 1970, Chap. 2, pp 47-77. (1957). (3) T.C.Rains."Atomic Absorption Spectroscopy," ASTM STP (6) George W. Snedecor and William C. Cochran, "Statistical Methods." 443, American Society for Testing and Materials, Philadelphia, Pa., 6th ed., The Iowa State University Press, Ames, Iowa, 1968, 591 1989. pp 19-36. pp.

ANALYTICAL CHEMISTRY, VOL. 45, NO. 8, JULY 1973 1511 L"'7

0.6 Table I. Example of Data Concentration 0.6 of added standards. Instrument response readings ppm Zn (absorbance) sampleAnalysis 0 (untreated) 0.197, 0.195, 0.199, 0.195. 0.197, 0.195, 0.197 0.5 0.291, 0.289, 0.289, 0.287, 0.293, 0.288, 0.287 1.0 0.381, 0.384, 0.383, 0.382, 0.384, 0.380, 0.384 2.0 0.556, 0.553, 0.551, 0.581, 0.559, 0.551, 0.553 n = 28; t9s = 2.056; Slope (absorbance/ppm Znl, Ms = 0.179. Ordi- nate Intercept (absorbance),ls = 0.199. Standard error of regression, In.tram.atcalibration (Sr)s = 13.81 X 10-0. Predicted value, IZa$ = 1.11 ± 4.98 X 10-2 at 95% C.L. Calibration line data: n = 217; t9s = 1.960; Slope (absorbance/ppm Zn), Mb = 0.183. Ordinate intercept (absorbance(, In = 0.102. Standard error of regression, (Sr)b = *4.16 X 10-3. Predicted value. IZbl = 0.559 ± 4.49 X 10 -2 at 95% C.L. Calculations of the values reported in the text were carried to more than three significant figures with the final reported values rounded to the third significant figure.

IIndiaat.. Data things 0.1 Table 11. Results of 12 Replicate Analyses by the Method of Standard Additions Extrapolated value, Estimated error, / i I 1 1 2.0 1.0 0 I0 2.0 3.0 4.0 Analysis No. ppm Zn Sz (Equation 8) 0.1.rmin.d I Addad 1 1.07 *1.92 X 10-2 CONCENTRATION, ArbitraryUnit. 2 1.11 ±1.14 X 10-2 Figure 1. Instrument calibration line and sample analysis line for 3 1.11 *1.48 X 10-2 the Method of Standard Additions. Zs and Zb are extrapolated 4 1.09 11.57 X 10 -2 values for the respective regression lines 5 1.08 ±1.31 X 10-2 6 1.09 ±1.88 X 10-2 7 1.09 *9.66 X 10-1 Standard error 8 1.11 ±2.42 X 10-2 ofregression, Sr = ±2;(Y -Y)1, (4a) 9 1.07 ±1.46 X 10-2 n-2 10 1.08 ±1.56 X 10-2 or in terms of the X and Y values, 11 1.10 *1.85 X 10-2 12 1.10 ±2.28 X 10-2 Sr ® ± EY2 - ('r Y)2 - (2;XY ZXnZYl2 Mean =1.09 Std dev =±1.47 X10-2 J(4b) ZX2 (EX)2 n n-2 Sz M A + B (8) where Y is the predicted value obtained from Equation 1 after M and I have been estimated. and a confidence level is given as From the theory concerning prediction in linear regression analysis (6-11)a confidence interval for the extrapo- ± tSz (9) latedline to the abscissa (designated as Z) by the method whereA= (n + 1)/n,B = (Z - X)2/E(X- X)2 =(D(Yo - of standardadditions is as follows: r)2/M2, Z on-I/M], tis the Studentt value forthe desired M(YC- Y) tCr +(nl confidence level with n - 2 degreesof freedom and the other + =k: D(Y, - Y)2 (5) symbols are as previouslydefined. Equation7 is valid only if M is highly significant, and if and the"best"single estimateof Z will be this conditionis notmet, Equation 5 should be applied. (yo Y Z X + M (6) EXPERIMENTAL Prepared Samples. In orderto test theprecisionof Z with ex- where}' is the arithmetic average of the Y values, Y. is perimentalvalues,a stock solution of zinc was prepared for anal- the value corresponding to the unknown value Z, M is the ysis. From this solution, twelve replicate analyses were made by regression coefficient (slope) in Equation 1, D = 1/T-(X - the method of standard additions, each consisting of an untreated X)2, t is the Student If value for the desired confidence and treated samples containing 0.5. 1.0, and 2.0 parts per million (ppml added zinc. A Perkin-Elmer Model 303 atomic absorption level with n - 2 degrees of freedom, and C = M2 - spectrophotometer and a digital concentration readout accessory t2Sr2D. (DCR-11 were used to measure the zinc concentration. Seven rep- If the slope M, as given by Equation 2 is highly signifi- licate readings per sample were performed giving a total of 28 cant, then C in Equation 5 approaches M2 and Equation 5 (7) American Society for Testing and Materials, Committee E-11 on reduces to quality control of materials, Philadelphia, Pa., Spec. Tech. Publ. 313. 1962, 28 pp. (Y - Y) tSr n + 1+ D(Y., - Y)2 (8) C. Eisenhart, Ann. Math. Statistics. 10. 162-186 (1939). + M2 (7) (9) Oscar Kempthorne and Leroy Folks, "Probability, Statistics, and bf + Al, n Data Analysis." list ed., The Iowa State University Press. Ames. Iowa, 1971. 555 pp. (10) Bernard Ostle. "Statistics in Research." 2nd ed., The Iowa State These limits are of the form Z ± tSz, where Sz denotes University Press. Ames. Iowa. 1963. 585 pp. the estimated standard error of Z for large samples, and (11) E. J. Williams, "Regression Analysis." John Wiley & Sons, New can be estimated by York, N. Y., 1959, 214 pp.

1512 ANALYTICAL CHEMISTRY, VOL. 45, NO. 8, JULY 1973 178

readings per analysis. (Inadvertently 3 of the samples ofthe entire analysis had one less reading.) Thedata from each analysis (absorbance Table Ill. Net Zinc Concentration with Estimated Error cs, concentration) were fittedto alinear least* Analy- squares model and the filled line was extrapolated to the abscissa. Net zinc Estimated error, sis No. concentration, ppm Table I illustrates data for a specific sampleanalysis. A calibra- Sd (Equation 12) tion line was established by reference to similarlyprepared stan- 1 5.10 X 10-' 13.38 X 10-2 dard solutions containing 0.5, 1.0, 2.0. and3.0 ppm zinc as well as 2 5.48 X 10-' 13.34 X 10-2 a blank solution. Before and periodically during thecourse of the 3 5.47 X 10-1 13.39 X 10-2 entire analysis, reference was made to thesestandard solutions 4 5.32 x 10-1 and blank to ensure proper calibration and 13.35 x 10-2 background readings 5 - 5.22 X 10-1 of 13.32 x 10-2 the instrument. Results of the estimatederror term for each 6 l 5.35 X 10-' 13. anaysis as given by Equation 8 from leastsquares regression 38 X 10-2 dat 7 5.34 X 10-1 13.33 X 10-2 a are shown in Table Il. These estimated uncertaintyterms show 8 5.53 X 10-' some variability, as expected, since in repeatedsampling 13.47 X 10-2 from th 9 - 5. e same universe, the uncertainty willvaryin width and 11X10-' 13.31X10-2 10 position fromsample to sample, particularly with small samples, 5.20X10-1 13.33X10-2 but should include ideally the true valueP percent (P is the se- 11 5.44 X 10'' 13.39 X 10-2 lected confidence level) of the time (7). Theprecision term, Sz, as 12 5.45 X 10-' 13.44 X 10-2 calculated from Equation 8 providesan estimate of the variabili- Mean = 5.33 X 10-' ty expected from sample to sampleas exemplified by the stan- 'S (sn - sbt = 12.85 X 10-2 dard deviati on of the twelve replicate analyses. "Known amount" = 5.0 X 10-' 11.0 X 10-2 Correcting for the Blank.Figure Iillustrates a sample analysis line along with the instrument calibration line. Theextrapo- For several Independent samples drawn from the same population with lated value for the sample analysis line (Zs) a given variance, the best estimate of the population variance is ob. must he corrected for tained_from the backgroundreading (/,b)since the instrument calibration line opting all the sample information. Thus.'Sis, v (Szn)t + 5612, where(Si) and (Sb) are calculatedfromEquationr a indicated the necessity of a blank correction.From the data of for the individual samples and the blank, and, theinstrument calibration line (Table 1) extrapolated to the ab- (Szp)2 = In, - 2) (Sz,)2 + In, - 2) (Szs)2.. +(112z 2) ISz,z)z scissa, a background reading of 0.559 ppni zincwas determined. (n,-21+(n2-2)+...+(nu-2) It is this valuewhich must lie subtracted from the extrapolated sample analysis in order to obtain thenet sample concentration. dicates a zinc concentration of 25 ± 3 (95% C.1..) Sinceth ppm. The e extrapolated background value and the sample analysis leaves were prepared for analysis by air-drying for 24 hours value areboth at 90 mean quantities and each has an estimated preci- °C and dissolving a weighed portion slowly in 70% HN0 sionterm, the 3on low difference between thesemeansalso will have an heat and the dissolved material diluted with O estimated ,IN HNO3.The uncertainty term. The variance of the differencebe- zinc calibration standards and blank were prepared tweentwoi in 0.I V ndependent means is the sum of their respectivevar- iances(i.e.,V HNO3. The leaves were analyzed by the previously described (Z.a - Zb) = V(Zs) + V(Zb)).The variance for an method ofstandardadditions. The zinc concentration determined extrapolatedvalue, Z, is estimated (from Equation 8) as follows: in the dried samplewas25.2 t 3.45 (95% C.L.) ppm, a range within the expected value. V(Z) °(M)2(A + B) (10) NOMENCLATURE When combining the variances of twomeans, a pooled standard X = concentration of standard solutions error of regression Sr is used (6), obtained from theregression Y = linearinstrumentresponse reading standarderror for both the instrument calibration line and the X = average of concentration values sample analysis line. =(2;X)/n f = average of instrument response readings= (E Y)/n Pooled standard error Y. = the Yvalue corresponding to the unknown Z of regression, f'= predicted Y value obtained from Equation1 after M and I have been estimated. SP , (n - 2) (Sr,)2 + (n5 - 2)(Sre)2 n = total number of readings made n, + nb - 4 (11) M = slope of regression line The estimated I= intercept of regressionline with oridinate axis precision term Sid, for the difference betweentwo Z = -1/M,I ZI = absolute value of Z independent means (Zs - Zb) is given by: A=(n+1)/n Sd - f (Sp) (A, + B,)+ (Aa + Br,) B(Z-X)2/Z(X-X)2 (12) C = M2 - t2Sr2D D=1/E(X-X)2 Subscripts a and b referto the sample and blank, respectively. A t = Student t value for appropriatedegreesof freedom confidence interval for Equation 12can be obtained by multiply. Sr =standard error of regression (Equation 4) ing by the appropriateStudent t value, for freedom. n + n - 4 degrees of Sp = pooled standard error of regression (Equation11) Table Sz = estimatederrorfor the sample extrapolated value Z III gives the net mean and estimatedprecision term as calculatedfrom Equation 12 for the values listed in fable (Equation 8) reeled for the instrument Il, cor- Sb = estimated error for the extrapolated background reading. It should beem- background phasized that the uncertainty interval may become increasingly value(Equation 8) large as thepredictedvalue moves Sd = the away from the mean (X) of estimated error term for the difference betweentwo standards, since estimates are expected tobecome poorer when independentmeans extrapolating away from the (Equation 12) range used in the original data (7). (Szp)2 = pooled values for allsample The error term, Sd, calculated from Equation12 provides information, Sz3 to mate of the an esti- Sz12 variability encountered between samplesas indicated by the standard error of difference forthe twelve replicate sam- V(Z) = variance ofZ = (Sz)2 (Equation 10) ples. Slsz(p)- se) = standard error of difference between two EnvironmentalAnalysis. The previously described was applied technique independent means for the pooled samplevalues and to an environmental sample consisting ofa quantity the background of dried orchard leaves obtained fromthe Nationala was B Standards(12) ureau of and containing certified concentrationsof several Received for review December 16. 1971. AcceptedJanuary elements. The Provisional certificate accompanying the leaves in- 29, 1973. This work supported by United StatesAtomic Energy Commission, contract number AT(45-1)2227 (12) W. Wayne Melnke, Anal. Chem.. 43(6),28A-47A Task (1971). Agreement 12.RLO-2227-T12-23.

ANALYTICAL CHEMISTRY, VOL. 45, NO. 8,JULY 1973 1513 RLO-2227-T12-24

FirstRecordsofBathysaurus mollis (Pisces: Bathysauridae) From the NortheastPacific Ocean

DAVID STEIN AND JOHN BUTLER Department of Oceanography Oregon State University, Corvallis, Oreg. 97331, USA

STEIN, D., AND J. BUTLER. 1972.First records of Bathysaurus mollis(Pisces: Bathysauridae) from the northeast Pacific Ocean.J. Fish. Res. Bd. Canada 29: 1093-1094. Three maleBathysaurus mollis,standard lengths 678, 458, and 440 mm, were captured 560 km off the coast of Oregon in 3358-3780 m. These are the first records of this species in the northeast Pacific. The 678-mm specimen, the largest reported individual of the species, has irregular rather than oval orbits. Variation exists between individuals in the position of the gill tooth plates.

STEIN, D., AND J. BUTLER.1972.First records ofBathysaurus mollis(Pisces: Bathysauridae) from the northeast Pacific Ocean.J. Fish. Res. Bd. Canada 29: 1093-1094. Trois specimens males deBathysaurus mollis,mesurant 678, 458 et 440 mm de longueur standard, ont ete captures a 560 km au large de la tote de l'Oregon, a des profondeurs de 3358 a 3780 m. Ce sont lespremieres mentions de cette espece dans le nord- est du Pacifique. Le specimen de 678 m, un record de longueur pour I'espece, a des orbites de forme irreguliere plutot qu'ovale. La position des plaques dentaires branchiales vane selon les individus.

Received January14, 1972

THREE maleBathysaurus mollis,the first of this215 and 188 mm SL. In our twolargest specimens, species recorded from the northeast Pacific, werethe number of gill tooth plates on the outside first captured with a 3-m beam trawl on Tufts Abyssalarch is 5 +14; in our smallest, 5 +16. plain about 560 km off Oregon by R/VYaquina.Two Only Mead (1966) has described the gill tooth individuals, 678 and 458 mm SL, were captured onplates in detail. Our specimens differ. Mead's (1966) May 31, 1970, at 44°55.9'N, 131°25.6'W, in 3358 m. specimens had one row of gill tooth plates located The third individual, 440 m SL, was captured onmesially on the first gill arches. In our smaller speci- June 5, 1970, at 44°32.0'N, 134°43.6'W, in 3780 m. mens, the positions of these plates are irregular; on These specimens were compared with all knownthe same arch they are paired in some places and reports ofB. mollis(Gunther 1878; Vaillant 1888;alternate from side to side in others. In our largest Vaillant, in Roule 1919; Townsend and Nichols 1925; specimen, the plates alternate regularly. These dif- Koefoed 1927; Nybelin 1957; and Mead 1966),ferences in gill tooth plate position may represent a with which they generally agree except for the gillchange with increasing size and age from small in- tooth plate counts. Mead (1966) reported 6 +15 gilldividuals with a single series of tooth plates to large tooth plates on the first gill arches of two specimensindividuals with alternating tooth plates. Our largest specimen (678 mm) is the largest re- Printed in Canada (J2401) ported for this species.It differs from all other 180

1094 JOURNALFISHERIESRESEARCH BOARD OF CANADA, VOL. 29, NO. 7, 1972

Bathysaurus mollishas been recorded throughout the Pacific Ocean: off Baja California in 3220 m, off Japan in 3429 m, in the South Pacific in 4361 m (Mead 1966), and now in the northeast Pacific in 3358-3780 m. It is also known from the northeast Outline of Orbit Atlantic in 2615-4360 m, and the Gulf of Mexico in 3150-4700 m (Mead 1966). The established range throughout the Pacific Oceansuggeststhat more sampling willshow thisspecies tobe very widespread.

Acknowledgments- The authorswould like to thank MrsM. M.Dick, Museum of Comparative Zoology, HarvardUniversity, for theloan of two B. mollis, and FIG. 1.Orbit of B. mollis. Dr W. G.Pearcy for his assistance and advice. This work was supportedby AEC Contract AT (45-1)2227 Task Agreement12, RLO-2227-T12-24. known specimens in having very irregular (Fig. 1) rather than oval orbits. The two anterior extensions of the orbit are skin covered and are not immediately GUNTHER, A.1878.Deep sea fishes collected during the voyage of the "Challenger".Ann. Mag. Natur. apparent. The larger of these extends forward from Hist. 5: 179-187. the anterior ventral quadrant, the smaller from the KOEFOED, E. 1927.Fishes from the sea-bottom.Rep. anterior dorsal quadrant. "Michael Sars" N. Atl. Deep-Sea Exped. 4: 1-148. In previous descriptions, B. mollishas been de- MEAD, G.1966.Bathysauridae.FishesWest.N. scribed as white (Townsend and Nichols 1925) Atl. 1: 103-113. or white with black bars (Mead 1966). The color NYBELIN, O.1957.Deep-sea bottomfishes.Rep. of our fresh specimens was white with a black patch Swed. Deep-Sea Exped. 2 (Zool.) 20: 250-345. of skin between the pectoral fins extending forward TOWNSEND, C., AND J.NICHOLS. 1925.Deep sea to the isthmus. The few scales remaining were in the fishes of the "Albatross" Lower California Expedition. samelocation, suggesting Bull. Amer. Mus. Natur. Hist. 52: 1-20. that skin and scales are VAILLANT, L.1888.Exp. Sci. "Travailleur" et "Talis- easily lost, and that the true coloration is black or man" poissons.Paris. 406 p. dark brown,as itisin another member of the 1919.In Roule. Res. Camp. Sci. Monaco 52: genus,B. agassizii,which is dark brown. 120-135.

I 1£1

RLO-2227-T12-25

Swimbladder Morphology and SpecificGravityof Myctophids off Oregon

Jom. L. BUTLER AND WILLIAM G. PEARCY Department of Oceanography Oregon State University, Corvallis, Ore. 97331, USA

BUTLER, J. L., AND W. G. PEARCY. 1972.Swimbladder morphology and specific gravity of myctophids off Oregon.J. Fish. Res. Bd. Canada 29: 1145-1150. Three general types of swimbladders were found in the eight species of myctophids studied: gas-filled, fat-invested, and atrophied or reduced. Small specimens of all species had thin-walled, gas-filled swimbladders. Large specimens of Stenobrachius leucopsarus had fat-invested swimbladders and large Diaphus theta had either gas-filled or atrophiedswimbladders,as found by otherworkers.Large Tarletonbeania crenularis had either gas-filled or reduced swimbladders, large Lampanyctus ritteri and L. regalis had reducedswimbladders,and large Stenobrachius nannochir had fat-investedswimbladders.Protomyctophum thompsoni and P. crockeri retained gas-filled swimbladders. High body lipid content was found in S. leucopsarus, S. nannochir, L. ritteri and D. theta, and low lipid content was found in the other four species. Myctophids with high lipid content had specific gravities close to that of sea water (1.026-1.030). Tarletonbeania crenularis with a reduced swimbladder had a specific gravity of 1.088. Lampanyctus regalis had a lower specific gravity(1.040)due to high water content of the tissue. The swimbladder to body volumes in S. leucopsarus and D. theta were inversely related to body size and lipidcontent,indicating that lipids assume the primary buoyancy function as the gas-filled swimbladder regresses with age. This change may eliminate the physiological constraints imposed by a gas-filled swimbladder and permit the more extensive diel vertical migrations of adults.

BUTLER, J. L., AND W. G. PEARCY. 1972.Swimbladder morphology and specific gravity of myctophids off Oregon.J. Fish. Res. Bd. Canada 29: 1145-1150. Nous avons identifie trois types generaux de vessie gazeuse chez huit esp8ces de Mycto- phidae: remplie de gaz, revetue de graisse et atrophies ou reduite. Les petits individus de toutes les especes ont une vessie 1 paroi mince, remplie de gaz. Les grands specimens de Stenobrachius leucopsarus ont une vessie gazeuse recouverte de graisse, et les grands Diaphus theta ontune vessie soit remplie de gaz, soft atrophies, comme l'ont remarque d'autres chercheurs. Les Tarletonbeania crenularisde grande taille ont une vessie remplie de gaz ou reduite,les grands Lampanyctus ritterietL. regalisont unevessie gazeusereduite, alors que les grandsStenobrac- hiusnannochiront une vessie gazeuse recouverte de graisse.Protomyctophum thompsoniet P. crockericonservent une vessie remplie de gaz. Nous avons observe une haute teneur en lipides chez S.leucopsarus, S. nannochir, L. ritterietD. theta,et une faible teneur dans les quatre autres esp8ces. Les Myctophidae A haute teneur en lipides ont un poids specifique qui se rapproche de celui de 1'eau de mer (1.026- 1.030).Tarletonbeania crenularis,qui a une vessie gazeuse reduite, a un poids specifique de 1.088.Lampanyctus regalisa un poids specifique plus faible (1.040) A cause de la haute teneur en eau de sestissus. Le volume de la vessie gazeuse par rapport A celui du corps est en relationinverse de la taille et de la teneur en lipides deS. leucopsaruset deD. theta,ce qui demontreque les lipides assument graduellement la fonction primaire de fiottabilite, A mesure que la vessie remplie de gazs'atrophie avec l'age. Ce changement peut eliminer les contraintes physiologiques qu'im- pose une vessie remplie de gaz, et permet des deplacements nycthemeraux plus etendus chez les adultes.

Received September14, 1971

Printed in Canada(J2297)

1145 182

1146 JOURNAI.FISIIERIESRESEARCII BOARD OF CANADA, VOL. 29, NO. 8, 1972

Myc ropuins have an euphysoclistous swimbladder To determine whether or not swimbladders contained which is often gas-filled (Marshall 1960). Because gas, fresh specimens were collected at night in mnid- they have gas inclusions that may effectively scatter water trawls in the upper 50 m. Fish were dissected under water and the swimbladder was punctured to and resonate sound and because they are common release any gas bubbles. The presence of gas in the oceanic animals that are known to undertake diet body cavity was considered evidence for gas in the vertical migrations, myctophids have been considered swimbladder since the swimbladder may have ruptured sources of scattering layers in the ocean (Marshall during the trawl ascent. One hundred thirty-nine speci- 1951, 1960; Tucker 1951; Hersey and Backus 1961; mens of eight species were examined in this manner. Taylor 1968). In addition, 206 frozen specimens were analyzed for The two most common myctophids (lanternfishes) lipid content after they were dried to a constant weight off Oregon,Stenobrachius leucopsarusandDiaphus at 70 C. Lipids were extracted with a two to one mixture theta,are known to migrate vertically within 12 kHz of chloroform and methanol in a Soxhlet extractor, scattering layer depths (Pearcy and Laurs 1966; then each specimen was dried and reweighed. Lipid content was calculated by difference. Small individuals Pearcy 1964; Pearcy and Mesecar 1971). These of the same size were grouped to provide a larger biomass. species were reported to lack a gas-filled swim-Seventy-eight extractions were made in this manner. bladder (Ray 1950; Jollie MS 1954; Barham MS Specific gravity was estimated by two methods. In 1956), but Capen (1967) found that small specimens the laboratory intact frozen specimens were weighed in ofS. leucopsarushad gas-filled swimbladders and air and in distilled water, and at sea fresh specimens suspected that small specimens ofD. thetaalso were suspended in a series of gum arabic solutions of may have gas in their swimbladders. Thus, adultsknown densities. Care was taken to exclude air from of these abundant species, lacking gas-filled swim- the gill chamber during all specific gravity measurements. bladders, may not scatter sound as effectively as juveniles of these species or other less common Typesof Swimbladders species with gas-filled swimbladders. This study was conducted on the morphology The myctophids were found to have three types of swimbladders to learn which species and sizes of swimbladders. Thin-walled, ellipsoid swimbladders (Fig. 1), were found in all small specimens of of myctophids off Oregon contain gas-filled swim-all species studied. As the fishes grow, the thin- bladders and which species lack gas-filled swim-walled swimbladder may be unmodified, or it may bladders. Such knowledge is needed to correlate become fat-invested in which case the gas phase is the depth distribution of effective sound scattering replaced by fatty tissue (Fig. 2), or the swimbladder organisms collected in nets with the depths of 3). sonic scattering layers. Lipids and specific gravity may be reduced in size and have thick walls (Fig. were also studied because of their interrelation The swimbladders ofStenobrachius leucopsarus andS. nannochirbecomes fat-invested with age. with swimbladders and buoyancy. Fatty reticular tissue on the anterior and posterior ends of the swimbladder in small specimens (Fig. 2) Materials and Methods stains red with Sudan IV in frozen sections indicating Fishes were collected off Oregon with an Isaacs-Kidd the presence of lipids. The reticular structure is midwater trawl equipped with a 5-mm mesh liner and a continuous with a layer of the swimbladder wall I in diameter codend of a 0.571 mm mesh. Preserved and is bounded by the outer layer, indicating that specimens were fixed at sea in 10% formalin and stored the fatty cells form inside the wall of the swim- in 36% isopropyl alcohol. Standard length was measured bladder, not outside as described by Capen (1967). to the nearest millimeter. As the fish grows larger, the fatty tissue increases Swimbladder volume was studied in 123 preserved and the size of the swimbladder is reduced until specimens of eight species of myctophids. The major and minor external axes of the swimbladder were mea- only a tube of fatty tissue remains in large indi- sured with an ocular micrometer and a stereo dissecting viduals. The volume of the entire swimbladder scope. For collapsed swimbladders, one half of the decreased from about 4-5% of total body volume circumference was measured. Total swimbladder volumes in small specimens of 20 mm to about 0.4% in were estimated by the formula for an oblate spheroid,large specimens (>70 mm) ofS. leucapsarus(Fig. V = 4/37rab2(Capen 1967), where a is the minor and 4A) and also inS. nannochir. b the major axis. Total body volume was estimated by The swimbladders ofLampanyctus ritteriand weighing individual fish in air and suspended in distilled L. regalisregress with increasing size of fish but, water, and estimating the volume of displaced water. spp., they do not become For study of their structure swimbladders selected unlikeStenobrachius from 46 fish were embedded in paraffin, sectioned, and invested with fat. The volume of the swimbladder stained with Mallory's triple stain or hemotoxin and decreased from about 2% of body volume in small eosin. Certain fat-invested swimbladders were sectioned individuals (<25 mm) to 0.01% or less in large with a cryostat and stained with Sudan IV for lipids. individuals (>80 mm). The swimbladder wall is 183

BUTLER AND PEARCY: MYCTOPHID SWIMBLADDER MORPHOLOGY 1147 thickened with connective tissue and the lumen is Gas was found in smallS.leucopsarusand occluded by the gas gland in large individuals. L. regalis,in both small and largeD. thetaand The swimbladders ofDiaphus thetado not grow T. crenularis,in largeP. thompsoni,and in small proportionally with the rest of the fish and are P. crockeri(Table 1). All of the gas-filled swim- sometimes reduced in large specimens. The swim-bladders had thin walls. Gas was never found in bladder volume in small individuals was aboutthe fat-invested swimbladders of largeS. leucopsarus 6-7% of the body volume (Fig. 4B). In large spec-or in the reduced swimbladders of large L.ritteri. imens (>25 mm) the swimbladder volume averagedSome of the largeT. crenularisandD. thetahad only 0.3% of body volume. There was considerablegas inclusions; but other individuals of both species variability in the large specimens, however. Somehad reduced swimbladders without gas. Thus, the had reduced, thick-walled swimbladders with thepresence of gas in thin-walled swimbladders and lumen occluded by the gas gland (Fig. 3), but otherthe absence of gas in fat-invested or reduced swim- adults of the same size had small, thin-walledbladders corroborated our findings based on the swimbladders that were not occluded. morphology of preserved specimens. The absence Swimbladders of largeTarletonbeania crenularis, of gas in largeS. nannochirandL. regalisand the likeD. theta,also have two distinct forms. Small,presence of gas in smallP. thompsoniand large reduced swimbladders were present in some spec-P. crockeriare inferred on the basis of the swim- imens while thin-walled, capacious swimbladdersbladder morphology and the presence or absence were present in others. The swimbladder to bodyof gas found in their congeners. volume was 0.5 % or less in reduced forms and greater than 6% in thin-walled forms. Thin-walled LypidContents and Specific Gravities swimbladders predominated (9 of 13) in specimens less than 40 mm; reduced swimbladders predom- Inlargeindividualsthe highestfatcontent inated (15 of 20) in larger specimens. occurred mainly in those species which lack a gas-filled swimbladder as adults, i.e.S. leucopsarus, All sizes of bothProtomyctophum thompsoniand S. nannochir,andL. ritteri;the lowest fat content P. crockerihad thin-walled swimbladders. However,occurred mainly in species which retain gas-filled estimated volumes of the swimbladders in bothswimbladders, i.e.P. thompsoniandP. crockeri. species varied, irrespective of size, from 5.5% toExceptions wereD. theta,which has either a small less than 1.0%0 of the body volume. The swimblad-gas-filled swimbladder or a reduced swimbladder, ders appeared collapsed in those specimens withbut a high fat content;T. crenularis,which has small volumes, but the lumen was not occludedeither a gas-filled or a reduced swimbladder, but by an enlarged gas gland as in the reduced swim-a low lipid content; andL. regalis,which has a bladders of the other species. reduced swimbladder and a low fat content.

TABLE 1.Occurrence of swimbladdergas infresh myctophids.

Gas present Gas absent

Species No. Length(mm) No. Length(mm)

Stenobrachius leucopsarus 27 Sa 22-39 5 Lb 48-72 S. nannochir S <35` L>35` Lampanyctus ritteri S <50c 8 L 50-120 L. regalis is <50 L> 50c Diaphus theta 11 S 17-22 2S 22-24 9L 34-62 30 L 29-70 Tarletonbeania crenularis 9S 23-33 5 S 29-34 4L 30-47 15 L 36-70 Protomyctophum thompsoni S <35C 13 L 35-50 P. crockeri 2S 19-26 L >30c

'Small individuals. "Large individuals. `Inferred from morphology. 184

L--, 0.1 mm

FIG. 1.(Top) Swimbladder of Diaphus theta, 23 mm standard length. g = gas gland, o = oval, r = rete mirabile. FIG. 2.(Center) Fat-invested swimbladder of Stenobrachius leucopsarus, 36 mm. f = fat-filled tissue, g = gas gland, r = rete mirabile, s = submucosa. FIG. 3.(Bottom) A reduced swimbladder of Diaphus theta, 38 mm. g = gas gland, s = submucosa, r = rete mirabile. Butter andPearcy-J. Fish. Res. Bd. Canada 185

Sr-

O`p Q a = N W

1 w I 31-

Vlie 10-

1 I

6 B iZ 5 00 Q ]

33W N4p 2

i I6 = I ti I .0-

N 3 ao R A h 3 4 W io

4 I I I ao 0 60 so STANDARD LENGTH Imm/

FIG. 4.Swimbladder volume and lipid content versus length of Stenobrachius leucopsarus (A)andDiaphus theta (B). 186

1148 JOURNAL FISHERIES RESEARCH BOARD OF CANADA, VOL. 29, NO. R, 1972

The total lipid contents of S. leucopsarus and Thawed specimens ofT. crenulariswere con- D. thetawere directly related to the size of thesiderably denser than sea water (1.088, Table 2), individual (Figs. 4A and B). The lipid contentas were seven fresh specimens withoutgas-filled of small individuals (<25 mm) was less than 10%swimbladders (1.086). The average specific gravity of of wet weight. Large individuals (>30 mm) hadeleven fresh specimens with gas-filled swimbladders a more variable lipid content which averaged 18.7was lower (1.056), but the expansion of gasduring and 22.1% of wet weight, respectively (Table 2).ascent from depth of capture may have produced Hence, lipid content and swimbladder volume were these low values. We conclude from these results inversely related in these species (Fig. 4A and B),that without a gas-filled swimbladderT. crenularis suggesting that lipids assume the buoyancy function would have to swim constantly to maintain position. of the gas-filled swimbladder as it regresses with age. Although the specific gravity of a fresh specimen The values for lipid content ofT. crenulariswereofL. regaliswith gas (50 mm) was less than 1.030, plotted against fish size, but the slope of a regressionlarge thawed specimens without gas (Table 2) were line for 22 specimens (22-70 mm) did not differdenser than seawater but not nearly as dense as significantly from zero (t = 0.4, 21 df). Moreover, T. crenularis.The body was flaccid and contained the mean lipid contents of fish with gas and fish86% water (on a weight basis) compared to 72% without gas were not significantly different (t = 0.4, inL. ritteri. Lampanyctus regalisis a lower meso- 15 df). Apparently, lipids do not have a major rolepelagic species (Pearcy 1964) that appears to have in buoyancy of this species even when the swimblad- achieved neutral buoyancy in the same way as der is reduced. bathypelagic fishes by reduction of skeletal and tissues(Denton andMarshall1958; S. leucopsarus, muscular The specific gravities of large Marshall 1960). S. nannochir, L. ritteri,andD. thetathat were frozen at sea and weighed in air and water were Discussion near the density of sea water (1.026-1.030) (Table 2). All the specimens of these sizes had fatinvested The swimbladders found in all juvenile mycto- or reduced swimbladders without a gas phase. Allphids, in the adults of P.thompsoni,andP. crockeri, of the specimens lost scales during capture soand in some adults of D.thetaandT. crenularis live animals may be denser than these results indicate.were thin-walled and gas-filled, morphologically Estimates of specific gravity of fresh specimenssimilar to those that Marshall (1960) characterized with gum arabic solutions were similar to thosefor the family. based on weights. The specific gravities of 17 fresh The situation inD. thetaandT. crenularis,in S. leucopsaruswith gas and three without gas werewhich both thin-walled and reduced swimbladders less than 1.030, and those of two specimens withoccurred, is curious, Capen (1967) also noted two gas and two without gas were between 1.030 andtypes of swimbladders inD. thetabut did not find 1.042. Those of three freshL. ritteriwere betweengas infresh specimens. We found no relation 1.030 and 1.048.Specific gravity estimates ofbetween type of swimbladder in these species and individuals which had gas-filled swimbladders areseason of the year or sex of individuals. Thus, of questionable value since gas could either expandthe ecological significance of this biphasic swim- during ascent from depths of capture or escapebladder system in the adults of these species is from ruptured swimbladders. unknown. Conceivably fishes with different swim-

TABLE 2.Average specific gravities of thawed myctophids frozen at sea. All individuals lacked gas-filled swimbladders.

Standard Species No. Length (mm) Specific Gravity

S. leucopsarus 5 58-80 1.027(1.025-1.031)a S. nannochir 2 90-102 1.026 L. ritteri 2 73-90 1.026(1.023-1.030) D. theta 4 40-64 1.038(1.025-1.062) L. regalis 2 84-85 1.040(1.040-1.041) T. crenularis 2 41-59 1.088(1.088-1.089)

aRange. 137

BUTLER AND PEARCY: MYCTOPHID SWIMBLADDER MORPHOLOGY 1149 bladder types could have different patterns of verticaloleyl oleate, a wax ester. Using this value for lipids migration and behavior. and 1.076 for fat-free tissue, a lipid content of Fat-invested swimbladders were found in large 19.4%0 of wet weight produces a specific gravity S. leucopsarus,confirming the earlier findings ofof 1.026. This value compares favorably with the Capen (1967), and inS. nannochir.Reduced swim-observed lipid contents (Table 2) forS. leucopsarus, bladders occurred in all largeL. ritteriandL. regalis S. nannochir, D. theta,andL. ritteri.Thus low- and in thoseD. thetaandT. crenulariswhich density lipids may provide neutral buoyancy for lacked thin-walled swimbladders. In general, bodythese species. Lee et al. (1971) found that trigly- lipid content was high in fishes that lacked gas-filledcerides were metabolized rapidly while wax esters swimbladders (with the exception ofL. regalis). were utilized more slowly in starvedGaussia princeps Similarly,Nevenzel etal.(1969) found larger(Scott), a large mesopelagic copepod. Accumulating amounts of lipids inL. ritteri, S. leucopsarus,andwax esters may have two advantages over trigly- D. thetathan inT. crenularis. cerides: a stable buoyancy fraction and a long-term Specific gravity measurements indicate that thoseenergy reserve. fishes with high lipid content(S. leucopsarus, S. The presence of a gas phase in swimbladders is nannochir, D. theta,and L.ritteri)approach neutralrelated to the vertical distribution and migration of buoyancy. Yet the observed lipid content is notmyctophids and their importance in producing as great as Taylor (1921) predicted for neutralsound scattering layers in the ocean. Of the four buoyancy. His theoretical value, based on specificcommon myctophids that Taylor (1968) caught in gravities of 1.076 for fat-free tissue and 0.925 for his study off British Columbia,S. leucopsarusand lipids (cod liver oil), is 29.23% of the wet weight. D. theta,species with fat-invested swimbladders Our highest observed lipid value, however, wasas adults, made the most extensive vertical mi- only 22. 1 % of wet weight (Table 3). Thus a lower grations and were most numerous below the main tissue density and/or lipid specific gravity is required sonic scattering layer during the day.Protomycto- to explain the low specific gravity of these fishes. phum thompsoni,a species with a gas-filled swim- The lipids found in myctophids may, in fact, bebladder, made less extensive migrations and was lighter than Taylor's value. Nevenzel et al. (1969)often found at scattering layer depths. Barham reported waxy esters (chain length 30-38) as the (1971) observed from submersible dives that im- principal lipids inS. leucopsarus, L. ritteri,andmature myctophids, with potentially resonant gas- D. theta.These lipids may be less dense than codfilled swimbladders, are sometimes concentrated liver oil, which is composed of lipids with chainat scattering layer depths while large individuals lengths of 18-20 (Morrison and Boyd 1966). Neven- were concentrated at deeper levels. Studies with zel et al. (1966) cite a specific gravity of 0.86 for opening-closing nets (W. G. Pearcy unpublished data) indicated that largeS. leucopsarus,without TABLE 3.Average lipid contentsof large individuals gas-filled swimbladders, have a greater vertical range of the various myctophids. of diel migration than juveniles, which have gas-filled swimbladders. However, smallS. leucopsaruswere Species No. also found in a 350-420 in scattering layer during °a Wet Wt. % Dry Wt. both day and night periods, indicating that some juveniles of this species probably migrate little if D. theta 28 22.1 ±0.7 63.6±1.2 at all. (Pearcy and Mesecar 1971). (> 35 mm) Maintaining a constant volume of gas in a swim- S. leucopsarus 28 18.7±4.0a 59.8±4.0 (> 30 mm) bladder during rapid, extensive vertical migrations is no mean physiological feat. It requires impressive S. nannochir 5 18.9±2.5 56.8±1.0 (> 85 mm) capabilities of gas secretion and resorption (Jones L. ritteri 10 16.0±0.3 57.8±2.0 1951, 1952; Kanwisher and Ebeling 1957; D'Aoust (> 58 mm) 1971; Alexander1971). These constraints may P. thompsoni 19 4.9±0.37 24.1±1.0 limit vertical movements of fishes with gas-filled (> 25 mm) swimbladders. Therefore replacement of gas by P. crockeri 2 4.2 19.2 lipids as the main buoyancy regulator, as in large (23 mm) S. leucopsarusandD. theta,may permit more exten- T. crenularis 25 4.2±0.2 19.2±0.8 (> 23 mm) sive vertical migrations in these myctophids. L. regalis 5 2.2±0.7 16.6±4.6 (> 50 mm) Acknowledgments This research was supported by the Office of Naval "One standard deviation of mean. Research through contract NOOO14-67-A-0369-0007 under 188

1150 JOURNAL FISHERIES RESEARCH BOARDOF CANADA, VOL. 29, NO. 8, 1972

project NR 083-102, andthe AtomicEnergyCommission KANWISHER, J., AND A. EBELING.1957.Composition ContractAT (45-1) 2227, TaskAgreement No. 12, of swimbladder gas in bathypelagic fishes.Deep-Sea RLO--2227-T12-25. Res. 4: 211-217. LEE, R. F., J. HIROTA, AND A. M. BARNETT.1971. Distribution and importance of waxesters in marine copepods and other zooplankton.Deep-Sea Res. ALEXANDER, R. M. 1971.Swimbladder gas secretion 18: 1147-1165. and energy expenditure in vertically migrating fishes. MARSHALL, N. B. 1951.Bathypelagic fishesas sound Proc. Int. Symp. on Biological Sound Scattering in scatterersin the ocean.J. Mar. Res. 10: 1-17. the Ocean. U.S. Dep. Navy, MC Rep. 005: 74-85. 1960.Swimbladder structure of deep-sea fishes in BARHAM, E. G. MS 1956.The ecology of sonic scat- relation to their systematics and biology.Discovery tering layers in the Monterey Bay area, California. Rep. 31: 1-122. Ph.D. thesis. Stanford Univ., Stanford, Calif. Univ. MORRISON, R. T., AND R. N. BOYD.1966..Organic Microfilms Publ. No. 21, 564; Ann Arbor, Mich. chemistry.Allyn and Bacon Inc., Boston. 1204 p. 1971.Deep-sea fishes: Lethargy and vertical orien- NEVENZEL, J. C., W. RODEGKER, J. F. MEAD, AND M. S. tation. Proc. Int. Symp. on Biological Sound Scattering GORDON.1966.Lipids of theliving coelacanth, in the Ocean. U.S. Dep. Navy, MC Rep. 005: 100-118. Latimera chalumnae.Science 152: 1753-1755. CAPEN, R. L.1967.Swimbladder morphology of NEVENZEL, J. C., W. RODEGKER, J. S. ROBINSON, AND some mesopelagic fishes in relation to sound scattering. M. KAYAMA.1969.The lipids of somelantern U.S. Navy Electron. Lab. Res. Rep. 29 p. fishes(family Myctophidae).J.Comp. Biochem. D'Aousr, B. G.1971.Physiological constraints on Physiol. 31: 25-36. vertical migration of mesopelagic fishes.Proc. Int. PEARCY, W. G.1964.Some distributional features of Symp. on Biological Sound Scattering in the Ocean. mesopelagicfishes off Oregon.J. Mar. Res. 22: U.S. Dep. Navy, MC Rep. 005: 86-99. 83-102. DENTON, E. J., AND N. B. MARSHALL.1958. The PEARCY, W. G., AND R. M. LAURS.1966.Vertical buoyancy of bathypelagic fishes without a gas-filled migration and distribution of mesopelagic fishes off swimbladder.J. Mar. Biol. Ass. U.K. 37: 753-69. Oregon.Deep-Sea Res. 13: 153-165. HERSEY, J. B., AND R. H. BACKUS.1954. New evidence PEARLY, W. G., AND R. S. MESECAR.1971.Scattering that migrating gas bubbles, probably the swimbladders layers and vertical distribution of oceanicanimals of fish are largely responsible for the scattering layers off Oregon.Proc.Int.Symp. Biological Sound on the continental rise south of New England.Deep- Scatteringin the Ocean. U.S. Dep. Navy, MC Rep. Sea Res. 1: 190-191. 005: 381-394. JOLLIE, M. T. MS 1954.The general anatomy ofRAY, D. L.1950.The peripheral nervous system of Lampanyctus leucopsarus(Eigenmann and Eigenmann). Lampanyctus leucopsarus.J. Morphol. 87: 61-178. Ph.D. thesis. Stanford Univ., Stanford, Calif. 239 p. TAYLOR, F. H. C. 1968.The relationship of midwater Univ. Microfilms Publ. No. 10, 376; Ann Arbor, trawl catches to sound scattering layers off the coast Mich. of northern British Columbia.J.Fish. Res. Bd. JONES, F. R. H.1951.The swimbladder and the Canada 25: 457-472. vertical movement of teleostean fishes.I. Physical TAYLOR, H. F.1921.Airbladder and specific gravity. factors. J. Exp.Biol. 28:553-566. Bull. U. S. Fish. 38: 121. 1952.The swimbladder and the vertical movement TUCKER, G. H.1951.Relation of fishes and other of teleostean fishes. 11.The restriction to rapid and organismsto the scattering of underwater sound. slow movements.J. Exp. Biol. 29: 94-109. J. Mar. Res. 10: 215-238. i.9

RLO-2227-T12-26 Reprinted from: InternationalAtomic EnergyAgency. 1973. Radioactive Contamination of the Marine Environment.[Proceedings of the Symposiumon the Interaction of Radioactive Contaminants of theMarineEnvironment,Seattle, Washington, 10-14 July 1972]. IAEA,Vienna.

ZINC-65 SPECIFIC ACTIVITIESFROM OREGON AND WASHINGTON CONTINENTAL SHELF SEDIMENTS AND BENTHIC INVERTEBRATE FAUNA

A.G. CAREY, Jr., N.H. CUTSHALL School of Oceanography, Oregon State University, Corvallis, Oregon, United States of America

Abstract

ZINC-65 SPECIFIC ACTIVITIES FROM OREGON AND WASHINGTON CONTINENTALSHELF SEDIMENTS AND BENTHICINVERTEBRATE FAUNA. Relationships between the benthic fauna and sediments on the continental shelf ofOregon and Washington have been investigated by determining specific activities of 65Zn.Zinc-65, induced by neutron activation in Columbia River water used to cool plutonium production reactorsat Richland, Washington, was carried down the river and into the marine environment.Sediment from the Columbia River moves northward. Zinc-65 specific activities in sediment and benthic invertebrates decrease northwardwith distance from the river and westward withbothdistance and depth. Many organisms to the south of the river mouth and westward with depth have higher 65Zn specific activities than the sedimentsin, or on, which they live. Other sources of 65Zn via the food web are suspectedas the cause for the difference in specific activity between the sediment and benthic fauna.Long-term declines in 65Zn specific activities have been noted. Many species from a range of trophic levels have been included in the study.

INTRODUCTION

The behavior and fate of radionuclides in the oceanic environment are the result of a complex of processes that includes the active concentration by organisms as well as physical and chemicalprocesses. Particularly for biologically active elements the effects of marine animals and plants must beconsidered. The sea floor as the lower boundary of the ocean is the site for many physical, chemical, and biologicalprocesses;it is a sink for many elements in theocean. Therefore,the sediments and associated organisms are portions of the marine ecosystem that play an important role in the fate of radionuclides in theoceans.The objective of this paper is to compare and contrast the distribution and concentration of 5Zn in the sedimentary environment and in the benthic faunaon the Oregon and Washington continentalshelf. We consider the specific activity of 65Zn (nCi 65Zn/gZn) in sediments to determine if the benthic invertebrate fauna living in, on or associated with the sea floorreflect the elemental composition of their sedimentary environment or whether other portions of the environment must be considered in the determination of theirsource of 65Zn.

The largest source of 65Zn to the area of the Northeast Pacific Ocean adjacent to the coasts of Oregon,Washington,and British Columbia has been until recently the plutonium production reactors at Hanford,Washington [1] [2]. Thereactors'cooling systems operation resulted in the neutron

287 190

288 CAREY andCUTSHALL

activation of traceelements in Columbia River water which passedthrough the reactors. Longer-lived radionuclides, including 65Zn (245-day half-life), were carried down the river and eventually into the Northeast Pacific Ocean. Some radionuclides were in particulate form or sorbed onto silt and clay particles and settled out in depositional areas in the river during their 600 km journey down the river [3].

Since 1967 approximately one plutonium production reactor has been permanently shut down each year; a total of eight have been phased out of operation since 1964. The last reactor utilizing the single-pass cooling system stopped operation in January1971. Becauseof the steady decrease in the number of reactors in operation, the input of 65Zn to the marine environment has been steadily decreasing.

The outflow of the Columbia River into the ocean, with its load of dissolved and particulate radionuclides, reacts to seasonal wind stress.Under the influence of northerly winds in the summer (approximately May through September), the river discharge flows to the southwest as a distinct lens- like plume of low-salinity water. During the winter period (approximately October through April) predominantly southerly winds push the Columbia River outflow to the north close to shore as a low-salinity surface layer; there it mixes with other sources of freshwater run-off. [4].

Zinc-65 from the Columbia River has been detected in all phases of the marine environment and in all ecological groups of marine biota. Radiozinc has been detected in phytoplankton, zooplankton, nekton, and benthos to depths of 2860 meters and to distances of 490 km off the central Oregon coast [5] [6] [7] [8]. It has been reported in sediment and water of the Northeast Pacific Ocean [4] [9] [10].

Many radionuclides in aquatic ecosystems become associated with sediments. Much of the Columbia River 65Zn appears in the sediments in the river and at sea [2] [3]. This paper explores relationships between the behavior of 65Zn in the sedimentary reservoir and the bottom-living organisms. As many of the benthic fauna significantly rework sediments in their search for food, their effect on the fate of 65Zn on the continental shelf will be considered. Benthos may be important in resuspending radionuclides in the water or introducing them into the oceanic food web. The results reported are part of a time series study of the decline of 65Zn in the marine environment of Oregon and Washington.

METHODS Sediments and benthic invertebrate animals were collected during 1970- 71 on the Oregon continental shelf adjacent to the Columbia River. The stations (Fig. 1) were patterned to sample a range of depths across the shelf at varying distances north and south of the river mouth. The trawling stations (Table I) were arranged in a series of transects normal to the coastline. Integrated cruises to sample both sediments and benthic fauna were scheduled in winter and summer to allow study of the effects of season and elapsed time on 65Zn specific activities.

The fauna was sampled by a 7 meter semi-balloon shrimp trawl with 12 ir.. (3.8 cm) stretch mesh. On the ship, the more abundant epibenthic invertebrates were sorted to species and deep-frozen in plastic bags. Care was taken to minimize trace metal contamination. The organisms were later dried for two weeks to constant weight in a forced draft drying oven at 65°C. For furtherconcentration of thesamples,the samples were ashed Al

IAEA-SM-158/17 289

48.6- 48

8 64 2 0I8 6 6 4 RRB Q W A S H

8 6 A 4 2 GRAYS 47 GH8 \ HARBOR 47-

WBB

19 46-

SEDIMENT STATIONS 6 JAN 1970 0 MAR 1970 0 DEC 1970 BEN TI-/OS STATIONS 45 45'

126 125 024 123

FIG. 1. Map showing sediment and benthic fauna station locations.The sampling date is indicated for each sediment sample; the standard trawling stations are indicated on the sampling transects (DIB, Destruction Island Benthos; RRB, Raft River Benthos.GHB, Grays Harbor Benthos; WBB, Willapa Bay Benthos; THB, Tillamook Head Benthos and NAD, Newport Anchor Dredge). in a muffle furnace at 450°C for 72 hours.The ash was ground to a fine powder in a SPEX mixer-mill and packed in 13 cm3 plastic tubes for gamma- ray analysis.Except for the larger organisms, several to several hundred individuals of a single species from each trawl were pooled to obtain enough ash. The samples were counted for gamma-emmissions for 400 or 800 minutes, depending on the radioactivity of the sample, in the well of a 12.7 x 12.7 cm sodium iodide (T1) crystal attached to a 512-channel pulse height analyzer. The radionuclide data were reduced by a least squares computer program TABLE 1. STATION LOACATIONS AND SAMPLING DATA FOR BENTHIC FAUNA w CCD

Location Distance Collection dates Depth from Station (m) Lat. Long river (km) - , o- r u .o 5 ur Destruction Island transact DIB 2 50 47'39.0'N 124'40.0'W 162.1-N X X DIB 4 100 47'39.0'N 124'50.0'W 166.1-N x X DI8 6 150 47'29.0'N 125'00.0'W 170.4-N x X DIB 8 200 47'39.0'N 125'05. 5'W 173.7-N X x

Raft River transect RRB 4 100 47'27, 0'N 124'37.6'W 143.5-N x x RRB 6 150 47'27, 0'N 124'45.8'W 143.9-N x

Grays Harbor transect

GHB 2 50 47'08.0'N 124'20.0'W 101.9-N GHB 4 100 47'08.0'N 124'32.0'W 103.3-N x x GHB 6 150 47'08.0'N 124.45.0'w 110.7 -N x x GHB 8 200 47'08.0'N 125'05.0'W 124.3-N x Willappa Bay transect WBB 1 30 46'38.0'N 124'10.0'W 43.5-N x x x WBB2 50 46'38.0'N 124'15.0'W 44.3-N x X x WBB3 75 46'38.0'N 124'18.2'W 42.2-N x WBB4 100 46'38.0'N 124'25. 0'W 50.0-N x x x WBB 6 150 46'38.0'N 124'31.0'W 54.4-N x x x WBB 8 200 46'38.0'N 124'38.0'W 60.2-N x x x Tillamook Head transect THB 2 50 45'56.0'N 124'02. 1'W 35.4-S x THB 3 75 45'56.0'N 124'06.0'W 35.2-S xx THB 4 100 45'56.0'N 124'12.2'W 36.7-S x THB 6 150 45'56. 0'N 124'28. 0'W 45.2-S x x x x THB 8 200 45'56.0'N 124'38.7'W 55.0-S x xx x x THB 19 2310 45'57.0'N 124'44.5'W 73.5-S x Newport transect NAD 2 50 44'39. 1'N 124'09.8'W 175.0-S x NAD 8 200 44'39. 1'N 124'36.0'W 179.3-S x xx 1v3

IAEA-SM-158/17 291

CARNIVOROUS ECHINODERMS 0 NOV 1969-APR 1970 0 Q MAY - OCT 1970

0 NOV 1970- APR 1971

A MAY - OCT 1971

DEPTH 100 200 6H8 NAD

FIG. 2. Zinc-65 specific activities for carnivorous echinoderms at depths of 25-200 m on the benthos station transects (DIB, Destruction Island Benthos; RRB, Raft RiverBenthos; GHB, Grays Harbor Benthos, WBB, Willapa Bay Benthos; THB, Tillamook Head Benthos; NAD Newport Anchor Dredge).Data points are for individual determinations with the period of collection indicated.Means for each station are plotted.

and reported as pCi/g ash-free dry weight. Data were corrected for decay during the time betweencollection and analysis.

Total zinc was analyzed by atomic absorption spectrophotometry. An aliquot of ash from eachsample wasprepared for element analysis by dissolving in concentratedHNO3and then diluting with 0.36N HC1.

Sedimentsamples wereobtained with a 0.1m2 spring-loaded Smith-McIntyre bottom grab [11]. The upper 1.0 cm (approximately 250 cm2) of the relatively undisturbed sections of the sedimentsample waslifted off. The sediment was immediately placed in 500 ml of 0.05 M CuSO, for 1-3 weeks to extract 65Zn and stableZn. The suspension was then filtered; the filtrate was analyzed for 65Zn by gamma-ray spectrometry and for Zn by atomic absorption spectrophotometry. Specific activities appear to be constant during the period of extraction and are not affected by uncertainties of extraction yields [2].

This procedure yields a higher specific activity for 65Zn in the extract than is obtained by total sediment dissolution or by boiling the sediment in aquaregia. Little can be said about the mode ofoccurence of Zn and 65Zninsediments, although it is clear that more thanone mode of binding is involved and that the specific activity of 65Zn in the dif-

ferent sediment "pools" is different. Leaching with CuSO4isa gentle form of extractionand producesa good yield. It is hypothetically possible that organisms extract 65Zn and Zn from a different pool than does CuSO,, and consequently extract a different specific activity than CuSO,,. For the present argumentswe assumethis is not the case. TABLE2.65Zn (pCi/g ash-free dry wt. ), STABLE Zn (pg/g ash-free drywt)AND 65ZnSPECIFIC ACTIVITY (nCi 65Zn/g Zn) IN BENTHIC INVERTEBRATES (These data appear in Figs 2, 3 and 4 in condensed form.Carnivorous asteroids are indicated by "predator")

Date of "Zn Specific Feeding Station Species collection ssZn Error Zn Error Activity Error type

(pci) 16 s. d. (pg) % s. d. (nCi ssZn/g Zn). s. d.

DIB 2 Pycnopodia helianthoides 10/21/71 32.67 3.16 256.0. 0.46 10.7 3.17 predator Paguridae 10/21/71 5.86 14.21 222.0 0.60 11.5 14,21 clan eon monitor 10/21/71 31.26 3.37 251.0 0.57 15.5 3; 38 ---

Pelecypoda 6/ 3/71 28.36 8.73 590.0 0.38 26.5 8.73 --- Luidit foliolata 6/ 3/71 24.64 7.43 314.0 0.72 6.5 7.44 predator DIB4 Craning, munita 10/21/71 32.67 3.84 266.0 0.83 13.8 3.86` --- Luidia foliolata 10/21/71 25.89 4.76 266.0 0.85 8.3 4.78 predator P coo..is helianthoides 10/21/71 32.81 3.96 262.0 0.97 9.9 3.99 predator Stylasteriasforreri 10/21/71 49.21 2.30 392.0 0.60 10.4 2.32 predator Metridium 10/21/71 86.50 2.52 812.0 0.45 10.6 2.53 ---

Mediaster aegualfs 6/ 3/71 10.73 20.53 74,3 1.70 11.9 20.55 --- Luidia foliolata 6/ 3/71 25.31. 7.42 336.0 0.60 5.7 7.43 predator Baltfcina actfica 6/ 3/71 25.37 7.50 101.0 0.91 19.5 7.51 ---

DIB 6 Goroaonocephalus caryi 10/20/71 1.93 3.40 239.0 0.95 0.8 3.43 predator Batticina pacific, 10/20/71 15.92. 5.94 130.0 1.35 9. 1 5.98 --- Brisasterlatifrom 10/20/71 3.18 27.88 36.6 5.50 7.2 28.02 --- DIB 8 Pandalus jordani 6/ 3/71 33.91 3.26 305.0 0.80 17.03.28 --- Allooentrotus fr 6/ 3/71 7.08 16.86 65.9 1.60 7.7 16.88 --- Luidiafoliolata. 6/ 3/71 17.04 6.76 317.0 0.45 4.7 6.76 predator Pandalus platyceros 6/ 8/71 40.42 4.91 329.0 0.54 11.24.92 Pandalus..lordanl 10/20/71 34.77 4.06 402.0 0.60 10.4 4.07 ---

Allocentrotus fragilis 10/20/71 3.68 23.11 76.5 1.20 3.5 23.12 --- Allocentrotus fraailis 10/20/71 4.52 17.73 143,0 1.00 2.9 17.14 Luidia foliolata 10/20/71 21.65 5.63 449.0 0.56 4.1 5.64 predator hiura sarsii 10/20/71 7.68 18.30 66.5 2.00 9.0 18.33 Gastropoda 10/20/11 11.85. 10.46 1607:0 0.10 9.0 10.46 --- TABLE 2.(cont.)

Date of 65 Zn Specific Feeding Station Species collection 65 Zn Error Zn Error activity Error type

- (pCi) Tr s. d. (jig) %s. d. (oCi ss Zn/g Zn) % s. d.

RRB4 Pandalus ordani 3/21/71 61.36 2.78 315.0 0.56 27.4 2.79 --- Luidia foliola 5 3/21/71 34.59 3.94 350.0 0.45 10.33.95 predator

Balticina pacifica 10/21/71 9.61 8.84 123.0 1. 15 9.8 8.86 --- Pycnopodia helianthoides 10/21/71 33.17 3.02 230.0 0.93 11.6 3.06 predator Usidia foliolata 10/21/71 38.15 2.69 419.0 0.52 7.4 2.70 predator

RRB 6 Munida guadrispina 3/21/11 26.25 7.81 128.0 1.50 22.17.85 --- Balticina pacifica 3/21/71 14.82 14.50 115.0 1.80 19.0 14.53 --- Gorgonocephalus casyi 3/21/71 40.58 6.99 165.0 0.91 21.6 1.00 predator

GHB 2 Metridium sp 10/22/71 179.82 0.95 1173.0 0.45 16.10.98 ---

GHB 4 Metridium sp 3/21/71 143,97 1.17 170.0 0.70 24.9 1.22 --- 1.89 Pandalus orl dani 3/21/71 83.7 1.81 315.0 1.10 31.4 --- Widsa foliolata 3/21/71 68.9 2.05 345.0 0.80 17.9 2.09 predator Cancer grac 3/21/71 48.11 6.09 161.0 1.10 30,7 6.11 ---

Pandalus jordant 10/22/71 41.02 2.48 396.0 0.50 13.5 2.49 --- Pycisopodia helianthoides 10/22/11 28.13 4.06 205.0 1.20 12.6 4.10 predator Gorgonocephalus caryi 10/22/71 39.44 2.60 213.0 0.72 16.4 2.62 predator

WBB 8 Pandalus 'alBanta 3/20/71 64.20 2.20 288.0 1.40 26.8 2.31 --- Munida quadrispina 3/20/71 29.40 4.14 113.0 2.80 23.9 4.31 --- Solaster endeca 3/20/71 6.18 15.51 49.1 3.30 9.7 15.60 predator

Pandalus jordani 10/24/71 40.65 4,77 378.0 0.50 13.9 4.78 --- Ophiuroidea 6/ 2/71 6.74 29.86 48.9 2.80 19.1 29.89 predator cno is helianthoides 6/ 2/71 16.59 12.82 161.0 0.69 9.5 12.82 --- Parastichopus californicus 6/ 2/71 19.30 12.42 157.0 0.95 12.1 12.43 --- Rossia pacifica 6/ 2/71 24.16 9.21 1197.0 0.30 30.39.21 ---

N Co W TABLE 2.(cont.)

THB 2 Luidia foliolata 9/25/70 82.56 2.00 286.0 2.62 24.12.39 predator THB 4 Gorgonocephalus caryi 3/20/71 67.80 4.88 238.0 0.56 22.8 4.89 predator Gotgonocephatus caryi 3/20/71 57.62 2.15 183.0 2.20 26.12.42 predator BalBctna pacifiCa 3/20/71 31.84 9.45 126,0 1.80 18,0 9.49 Balticina cifica 3/20/71 22.42 4.71 87.3 3.30 18.34.99 Panda Ws or] dani 3/20/71 52.76 2.27 299,0 1.30 33.32.36 Luidia foliolata 3/20/71 59.40 2.32 288.0 0.60 15.7 2.34 predator Gorgopocephalus caryi 3/20/71 67.80 4.88 238.0 0.56 22.8 4.89 predator

Balticina cifica 6/ 1/71 34.37 5.15 140.0 1,10 20.8 5.18 Pandalus lOrdaoi 6/ 1/71 64.23 3.08 265.0 0.98 28.8 3.12 Luidla foliolata 6/ 1/71 16.18 8.68 117.0 1.50 30.18.71 predator Metridium 6/ 1/71 5.49 29.68 280.0 1.10 27.0 29.69

THB 6 Pentamera poputifera 12/ 8/70 13.20 8.61 62.6 5.48 28.09.04 Brisaster latifrons 12/ 8/70 6.79 17.91 18.9 11.80 30.0 18.86

Pandalus jordani 6/ 2/71 55.89 2.13 323.0 0.79 22.12.76 Isopoda 6/ 2/71 9.57 13.55 701.0 0.64 22.0 13.55

Pandalus asl dani 10/25/71 45.44 2.99 300.8 0.86 17.33.02 Pentamera populifera 10/25/71 9.88 15.99 57.8 2.86 14.6 16.05

THB 8 Luidia fotiolata 1/ 3/70 26.51 4.90 1314;0 0.40 13.04.90 predator Pandalus pcdani 1/ 3/70 16.48 7.42 332.0 1.11 45,1 1,44

Allocentrotus fraallis 3/21/70 5.06 22.25 53.8 5.73 8.9 22.43 Rathbunaster californtcus 3/21/70 21.21 7.58 219.0 1.63 17.2 7,62

Pandaausltrdwi 12/ 8/10 58.08 2.64 352.0 0.89 40.4 2.68 Rathbunaster californicus 12/ 8/70 43.34 3.37 319.0 1.19 27,5 3.42 predator Pseudarchaster 12/ 8/70 4.86 13.56 73.3 4.47 8.9 13.74 predator Luidta foltotata 12/ 8/70 33.29 3,79 265.0 1.79 42.2 3.89 predator Brisaster latifrons 12/ 8/70 7.69 16.08 51.9 6.16 22.8 16.37 Parastichopus caltfomicus 12/ 8/70 30.29 4.52 88.6 8.89 32.6 6.34

Parastichopus cattfornicus 9/26/70 60.57 4.55 172.0 1.20 31.6 4.59 Pandalus jordani 9/26/70 68,80 3.98 420.0 0.70 38.14.00 Rathbunaster californtcus 9/26/70 43.05 5.96 231.0 1.70 16.7 6.02 predator ------TABLE 2.(cont.)

Date of ssZn Specific Feeding Station Species collection s5Zn Error Zn Error activity Error type

(pCi) tos.d. (4g) 'ro s. d. (nCi ssZn/g Zn) ( s, d.

Pandalus rlo dani 2/13/71 12.14 12.29 269.0 1.20 17.6 12.30 --- Parastlchopus californicus 2/13/71 34.61 5.23 137.0 2.50 19.4 5.38 --- Crossaster 26220 9 2/13/71 6.69 21.48 132.0 0.90 14.1 21.48 predator Lopholitholdes foraminatus 2/13/71 54.25 3.29 886.0 0.60 14.6 3.30 --- Lopholltholdes foraminatus 2/13/71 65.10 2.95 3535.0 0.50 14.0 2. 96 --- Pandaiusiarrdani 3/19/71 43.61 2.76 246.0 2.50 23.5 3.03 --- Pseudarchaster parch! 3/19/71 5.73 16.67 88.2 3.30 5.5 16.75 predator Aphrodite sp 3/19/71 32.86 3.42 447.0 1.90 13.5 3.55 --- GHB 6 Pandalus dani 3/21/71 70.10 2.75 324.0 0.50 30.02.76 --- Munida guadriapina 3/21/71 13.75 11.25 176.0 0.65 20.4 11.25 --- Gorgonocephalus caryi 3/21/71 29.39 5.12 184.0 0.80 19.4 5.14 predator

Munida quadrispina 10/22/71 14.76 6.23 128.0 1.10 10.5 6.25 --- Octopoda 10/22/71 14.14 6.50 2056.0 0.35 9.6 6.50 ---

GHB 8 Pandalusor1dani 3/21/71 36.87 4.18 261.0 0.8 23.04.20 --- Allocentrotus fragilis 3/21/71 15.42 9.89 427.0 0.5 9.2 9.89 --- Allocentrotus fragilis 3/21/71 5.77 25.25 64.0 2.4 6.5 25.28 Hippasterias isp nosy 3/21/71 7.52 17.86 65.2 1.8 11.3 17.88 predator

WBB 2 Cancer ma ter 3/20/71 130.00 1.30 300.0 2.30 39.11.74 --- Crangon manila 10/24/71 30.25 4.16 249.0 1.10 24.9 4.20 --- Luidia foliolata 6/ 2/71 82.72 2.07 293.0 0.40 22.6 2.08 predator

WBB 4 Luidia foliolata 3/20/71 85.48 1.60 285.0 1.10 22.9 1.69 predator Pycnopodia helianthoides 3/20/71 43.46 2.70 190.0 1.80 20.9 2.85 predator Pandalusorldani 3/20/71 44.96 2.66 311.0 1.40 31.22.75 -- Pandalus ordanf 6/ 2/71 74.50 4.06 210.0 0.86 29.04.08 --- Pycnopodia helianthoides 6/ 2/71 44.13 3.93 237.0. 0.73 16.6 3.95 predator Gorgonocephalus caryi 6/ 2/71 45.20 5.30 178.0 1.50 25.8 5.35 predator

------N ctD TABLE 2.(cont.)

Rossla pacifica 6/ 2/71 14.47 14.62 897.0 0.33 34.1 14.62 --- Balticina cifica 6/ 2/71 37.82 7.09 164.0 0.83 26.8 7.10 --- Metridium ssp 6/ 2/71 174.71 2.16 688.0 0.48 22.6 2.17 --- Paguridae 6/ 2/71 12.25 18,40 241.0 0.68 16.6 18.40 ---

Gorgonocephalus caryi 10/24/71 45.36 3.28 201.0 1.10 11.4 3.33 predator Pandalus jordani 10/24/71 44.58 3.26 328.0 1.30 18.6 3.32 --- Luidia foliolata 10/24/71 38.34 2.60 227.0 0.81 14.8 2.63 predator Crangon mun(ta 10/24/71 19.51 6.43 212.0 0.77 21.4 6.44

WBB 6 Parastichopus califomicus 12/ 9/70 39.16 3.70 176.0 2.30 33.9 3.87 --- Paraniopsis inflata 12/ 9/70 19.11 4.13 107.0 2.98 22.5 4.39 --- Hippasterias spi¢osa 12/ 9/70 5.66 21.20 42.5 7.14 30.6 21.50 predator Mundia quadrispina 12/ 9/70 23.58 5.69 131.0 5.28 24.7 6.27 --- Pandalus jordani 12/ 9/70 47.08 3.13 359.0 1.69 31.5 3.24 --- hiura sarsii 12/ 9/10 6.26 19.44 40.2 8.94 42.5 19.95 --- Pteraster tesselatus 12/ 9/70 46.02 3. 18 366.0 1. 12 19.7 3.23 predator Mediaster aequalis 12/ 9/70 10.11 10.94 63.0 4.69 19.4 11. 19 ---

CrangO9 franciscmum 3/20/71 58.93 2.16 317.0 0.80 24.8 2.20 --- Pandalus jordani 3/20/71 77.37 1.93 314.0 1.50 30.21.98 --- Octopoda 3/20/71 72.30 2.01 1263.0 0.40 24.5 2.02 ---

Pandalus joedani 10/24/71 50.03 3.86 364.0 0.52 16.23.87 --- Munida quadrispina 10/24/71 13.57 12.13 94.8 2.10 11.9 12. 18 --- ` Pseudarchaster pmelii 10/24/71 7.74 20.16 69.8 3.20 8.6 20.22 predator Stylasterias forreri 10/24/71 31.30 3.97 277.0 0.88 7.3 3.99 predator

Pandalusjordani 10/25/71 43.71 3.34 341.0 0.46 14.23.35 --- Rathbunaster califomicus 10/25/71 10.17 11.24 206.0 1.10 4.9 11.25 predator AUocentrotus fragilis 10/25/71 6.91 16.42 60.2 4.30 6.2 16.56

NAD 2 Pilaster brev,inns 9/28/71 21.19 6.64 131.0 2.14 12.7 6.73 predator

NAD 8 Luidia foliolata 1/ 3/70 17,53 7.11 122.0 3.64 13.07.34 predator Rathbunaster californicus 1/ 3/70 40.38 6.00 298.0 1.50 12,7 6.05 predator is 9

IAEA-SM-158/17 297

RESULTS

General Trends in Faunal 65Zn Specific Activities

The data in Fig. 2 are derived from carnivorous echinoderms (asteroids). Zinc-65 specific activities (nCi 65Zn/gZn) exhibit a general pattern associated with distance from the Columbia River mouth and with depth. Specific activities at station transect lines further from the river at a distance of 160-180 km north and south are low (Fig. 2 and Table 2), while those from stations adjacent to the river are highest. Less radiozinc has decayed during the short transit time to nearby sites and there is less chance for isotopic dilution. Specific activities for 65Zn generally decrease with depth and distance from shore presumably because of increased vertical and horizontal transport time. These results have been noted elsewhere for 65Zn concentrations and specific activities in benthic invertebrates in the North- east Pacific Ocean [5] [8]. No general trend with season and year has been detected for these data.

Trophic level can effect the levels of 65Zn and consequently the specific activities. There is an inverse relationship between trophic level and 65Zn specific activities [5] [7] [8] [12]. Data from organisms at one trophic level have been utilized in this analysis to demonstrate general trends in the faunal specific activities within the study area. Predatory asteroids are ubiquitous on the continental shelf in the Northeast Pacific Ocean [13], and their food source is generally known [14]. They feed at the same trophic level, as they preyuponsediment-detrital feedingorganisms.

Effect of Depth on 65Zn Specific Activities Specific activites for 65Zn in sediments and benthic epifauna change markedly with increasing depth and distance from The Columbia River (Fig. 3 and Table 2).Though the general trend is a decrease in specific activity with depth, the fi5Zn concentrations in the benthos and in their sedimentary environment differ over a large portion of the continental shelf. The specific activity in surface sediments are at intermediate levels at 25-50 m depth, increase to a maximum at 80-100 m, and then decline with depth. Zinc-65 cannot be detected in sediments on the Tillamook Head transect on the continental slope at 950 m depth or below. Benthic organisms contain the most 65Zn inshore at the shallowest stations sampled (25-30 m).Animals collected on the same dates as the sediment samples in December 1970 and March 1971 have higher 65Zn specific activities at these inshore stations than those found in sediment extracts. The sedimentary 65Zn has higher specific activities at mid-shelf depths (60-100 m), but from 150 m down to 2370 m depth (station THB 19 on the Tillamook Headtransect),the benthic invertebrates consistently have higher specific activities for zinc-65 than do the sediments. Zinc-65 can be detected in bottom-dwelling organisms beyond 2000 m whereas in the sediments it is below our detection limits.

These data indicate that 65Zn reaches the benthic fauna along other pathways insteadof,or in addition to, the sedimentarysource. Low levels of 65Zn have been reported for Willapa Bay sediments [10], while oysters from the bay have accumulated higher concentrations [9]. These filter- feeding members of the benthos probably obtain 65Zn from particles in the water rather than from bottom sediments. Zinc-65 is at low levels in the nearshore sediments where there are significant tidal and wave venerated bottom currents to keep the finer particles in suspension. As 5Zn can be 200

298 CAREYand CUTSHALL

WILLAPA BAY TRANSECT SEDIMENT - January 1970 A March 1970 II°4` 0° 0 December 1970

501- March 1971 I

WILLAPA 13AY TRANSECT EPIBENTHIC INVERTEBRATES 0 - 0 December 1970 March 1971 11--6 June e1971 -.October 1971

360 .400 440 460r 2400 40 60 120 160 200 240 260 320 DEPTH (m) FIG. 3. Zinc-65 specific activities for benthic epifaunal invertebrates and for sediment from the Willapa Bay transect (WBB). Means t 1 sample standard deviation are plotted. The number of observations is indicated above each point.The station at 2370 m depth is THB 19 on the Tillamook Head transect (45' 56.8' -N, 125' 29' -W).Animals analysed include a broad range of taxa and ecological types.

detected in bathyal benthic invertebrates (Fig. 2) [2], it is evident that the fauna is concentrating zinc. The detrital food web in the food-poor abyssal environment may be the primary pathway of zinc-65 to the benthic fauna.

Effect of Distance on 65Zn Specific Activities The highest 65Zn specific activities are found in benthic organisms and sediments in the depositional area directly west of the river (Fig. 2, 20.

IAEA-SM-158/17 299

So,- 30

7q-

601-

50 - INNER SHELF 0 SEDIMENT Jan7O-MOr7O(37-65m) BENTHIC INVERTEBRATES 401- Dec 70-Oct 71 (50 m)

3OH 0

20F- 0 H- 43-

ct I I 1 I 1 1 I 1 I I

50r- 4 MID-SHELF 0 SEDIMENT Jon 70-Mor 70 (86-118m) e0H BENTHIC INVERTEBRATES Dec 70-Oct 71 (1OOm)

301-

201-

IOF 43a

so, OUTER SHELF 0 SEDIMENT ion 70-Mor 70 (160-202-) 40t- BENTHIC INVERTEBRATES Dec 70-Oct 71 (200 m)

32 30F_

33 20-

I I INOIITHI KO 120 e0 40 120 160 (SOUTH) DISTANCE FROM COLOMBIA RIVER /Am) FIG. 4. Zinc-65 specific activities for sediment and epibenthic fauna along 3 depth contours:inner continental shelf, mid-shelf, and outer shelf from the Destruction Island Transect (DI8) to the Newport Transect (NAD). Means * I sample standard deviation are plotted.When more than one observation was made, the number is indicated.Animals analysed include a broad range of taxa and ecological types. 2 02

300 CAREY andCUTSHALL

Fig. 4, and Table 2) on the inner continental shelf. The levels decrease to the north and south of the river mouth. This pattern is evident in deeper water, but the amplitude is damped with increasing depth and distance from the river. The specific activities in sediments at 100 m depth west of the river mouth are about one half those at 50 m depth, while benthos specific activities decrease only slightly. Further west at the edge of the continental shelf at 200 m depth, sediment specific activities are again significantly decreased while the faunal specific activities are much less so.

At the mid and outer shelf environments, the 65Zn in the sediments to the south of the river have low to zero specific activities. The distribution of 65Zn predominantly to the north of the Columbia River may be caused by the long duration of northerly flowing surface currents during the fall, winter and early spring months [10]. Seabed drifter data suggest that northward flow of water is common near the bottom throughout the year over the continental shelf especially at depths greater than 40 m [15].

Effect of Time on 65Zn Specific Activity

The decreasing input of 65Zn into the marine environment since December 1964 is reflected in specific activity levels in the sediments [2] and in the benthic fauna [12] [16]. Specific activities in continental shelf

SUMMER I WINTER WINTER SUMMER I WINTER I SUMMER I WINTER I SUMMER I WINTER I

901- 0

O PARASTICHOPUS CALIFORNICUs O THB 8 (200m) NAD 8 (200m)

0 0

0 0 0

Jon 67 10 20 Jon 69 30 J" 70 40 TIME (cumulative in months)

FIG. 5. Zinc-65 specific activities for the holothurian, Parastichopus californicus from April 1967 to February 1971 from 200 meters depth on the Tillamook Head Transect (THB) and Newport Transect (NAD). 2203

IAEA-SM-158/17 301

TABLE 3.SPECIFIC ACTIVITY OF 65Zn IN Parastichopus californicus (Echinodermata, Holothuroidea) FROM JULY 1967 TO FEBRUARY AND MARCH 1971 FROM 200 m DEPTH ON THE TILLAMOOK HEAD (THB) TRANSECT AND NEWPORT (NAD) TRANSECT

Date of 65Zn Specific Collection bSZn Error Zn Error Activity Error 65 (pCi (o s.d.) (ugZn/ (% s.d.) (nCi 5Zn/g Zn)(% s.d.) g ash-free g ash-free dry wt.) dry wt.)

THB 8

7/27/67 123.99 5.47 148.0 7.4 83.0 6.60 1/11/68 71.13 9.81 286.4 1.4 48.5 9.83 4/ 1/68 37.29 8.41 113.4 1.9 38.38.46 10/13/68 59.06 5.78 120.0 7.0 89.9 6.76 7/13/69 49.82 6.27 91.1 3.2 61.1 6.47 10/14/69 91.92 12.81 113.0 6.5 63.3 13.22 1/ 3/70 44.45 6.33 183.0 4.3 22.16.69 12/ 8/70 30.29 4.52 88.6 8.8 32.6 6.34 2/13/71 34.61 5.23 137.0 2.5 19.4 5.38

NAD 8

7/24/67 24.26 12.0 148.0 2.7 37.0 12.08 7/24/67 6.13 29.57 37.0 2.7 26.9 29.60 7/24/67 60.82 6.0 168.9 2.3 11.6 6.11 4/ 4/68 37.28 7.94 108.0 3.1 37.8 8.09 4/ 4/68 22.78 13.78 126.0 2.7 35.1 13.85 1/ 3/69 28.91 13.19 110.0 1.0 23.5 13.20 11/20/69 16.48 8.68 115.0 4.6 11.38.98 2/18/70 14.97 8.54 134.0 2.4 8.6 8.62 3/ 9/70 14.25 9.46 129.0 5.1 9.1 9.80 3/ 9/70 9.07 15.67 50.6 3.7 12.7 15.78 5/27/70 20.67 7.10 136.0 2.3 11.9 7.19 9/16/70 22.97 5.00 230.0 1.4 8.9 5.05 9/28/70 17.49 7.64 142.0 2.1 10.7 7.71

3/19/70 7.25 19.76 130.0 1.3 4.3 19.77 204

302 CAREYand CUTSHALL

benthicfauna,asexemplified bytheholothurian, Parastichopuscalifornicus, have decreased aboutfive-fold at theedgeof the continentalshelf west of Tillamook Head and Newport(Fig. 5 and Table3).Parastichopus feeds by sweeping the sedimentsurface withmucous-covered tentacles 17;it should be representativeof detrital feeders. Thespecificactivityin this species is loweroff Newport,179 km south of the Columbia Rivermouth,and the decline is moreregular. The large fluctuationsof 65Zn specific activity in organismsoff TillamookHead is thought to be caused by marked seasonal changes in the position of the Columbia River plume.

CONCLUSIONS AND SUMMARY

1. Zinc-65 specific activities are highest in benthic fauna and sediments on the inner continental shelf directly west of the Columbia River mouth.

2. Beyond the depositional area at the river mouth zinc-65 specific activities in sediments reach a maximum on the mid continental shelf at depths of 60-100 m.

3. Zinc-65 specific activities in fauna and sediment generally decrease with depth (and transporttime),though the specific activities are higher in fauna than in the sediments on the continental slope and deeper environments beyond. 4. Specific activity for 65Zn in benthic fauna and sediments decreases with distance (and transporttime)from the ColumbiaRiver. Specific activities in sediment reach zero about 80 km south of the river while organisms 100 km further to the south off Newport still contain significant amounts of 65Zn. 5. The Columbia River plume transports zinc-65 to benthic organisms to the south of the river and probably to the north also.

6. Specific activities for 65Zn have declined five-fold in five years. The levels of specific activities in faunaoff TillamookHead close to the riverhave significant seasonal variations. 7. Benthic organisms are often not in equilibrium with the 65Zn specific activities of their sedimentaryenvironment. They probably acquire 65Zn with differingspecific activities from other sources such as the detrital food web.

ACKNOWLEDGEMENTS

We particularly thank K. E. Nafe,R.E. Ruff, C. L. Kuelthau, M. A. i(yte, G.Wagner,and I.L.Larsen for their invaluable assistance on board ship and for the many long hours involved with preparing and analyzing samples.We thank the U. S. Atomic Energy Commission for financial support (Contract No. AT (45-1) - 2227, Task Agreement 12; RLO-2227-T12-26). 2(5

LAEA-SM-158/17 303

REFERENCES [1] OSTERBERG, C. L., CUTSHALL, N., JOHNSON, V., CRONIN, J., JENNINGS, D. and FREDERICK L. Non-biological aspects of Columbia River radioactivity, Disposal of Radioactive Wastes into Seas, Oceans, and Surface Waters, IAEA, Vienna (1966) 321.

[2] CUTSHALL, N., W. C. RENFRO, EVANS, D. W., JOHNSON, V. G. Zinc-65 in Oregon-Washington Continental Shelf Sediments, Third Radioecology Symp. (Proc. Conf. Oak Ridge, in press).

[3] PERKINS, R. W., NELSON, J. L., HAUSHILD, W. L. Behavior and Transport of Radionuclides in the ColumbiaRiver betweenHanford andVancouver, Washington, Limnol. Oceanogr. 11 2(1966) 235.

[4] BARNES, C. A. and GROSS, M. G. Distribution at Sea of Columbia River water and its load of radionuclides, Disposal of Radioactive Wastes into Seas, Oceans and Surface Waters, IAEA, Vienna (1966) 291.

[5] CAREY, A. G., Jr., PEARCY, W. G., OSTERBERG, C. L., Artificial radionuclides in marine organismsin the Northeast Pacific Ocean off Oregon, Disposal of Radioactive Wastes into Seas, Oceans, and Surface Waters IAEA, Vienna (1966) 303.

[6] PEARCY, W. G., OSTERBERG, C. L. Depth,diel, seasonaland geographic variations in zinc-65 of midwater animals of Oregon, Int. J. Oceanol. and Limnol. 1 2(1967) 103.

[7] OSTERBERG, C. L., PEARCY, W. G., CURL, H. C., Jr. Radioactivity and its relationship to oceanic food chains, J. Mar. Res. 22 1(1964)2.

[8] CAREY, A. G., Jr., 65Zn in benthic invertebrates off the Oregon coast. Columbia River Estuary and AdjacentOceanWaters: Bioenvironmental Studies. PRUTER, A.F., ALVERSON, D. L. eds.Univer. Washington Press, Seattle. (1972) 833.

[9] SEYMOUR, A. H., LEWIS,G. B. Radionuclidesof Columbia River origin in marineorganisms,sedimentsand water collected from the coastal and offshore waters ofWashingtonand Oregon, 1961-1963, U. S. Atomic Energy Comm. UWFL-86 (1964).

[10] GROSS, M. G. Distribution of radioactivemarine sediment derived from the Columbia River, J. Geophysical Res. 71 8(1966)2017.

[11] SMITH, W., McINTYRE, A. D. Aspring-loaded bottom sampler. J. of Mar. Biol. Assoc. 33 1(1954) 257.

[12] PEARCY, W. G., VANDERPLOEG, H. A.Radioecology of benthic fishes off Oregon, U. S. A. The Interaction of RadioactiveContaminants with the Constituents of the Marine Environment, IAEA,Seattle (in press).

[13] ALTON, M. S. Bathymetric distributionof sea stars (Asteroidea) off the northernOregoncoast, J. Fish.Res.Bd. Canada 23 11(1966) 1673. [14] CAREY, A. G., Jr. Food sources of sublittoral, bathyaland abyssal asteroids in the Northeast Pacific Ocean.Ophelia 10 l(in press).

[15] GROSS,M. G.,MORSE,B., BARNES, C. A. Movementof near bottom waters on the continental shelf off the northwestern United States. J. Geo- physical Res. 74 28(1969) 7044. 206

304 CARER and CUTSHALL

[16] RENFRO, W. C. Seasonal radionuclide inventories in Alder Slough, an ecosystem in the Columbia River estuary. Second Nat'l Symp. on Radioecology (Proc. Conf. Oak Ridge, in press).

[17] MACGINITIE, G. E., MACGINITIE, N. Natural History of Marine Animals McGraw-Hill, New York (1949).

DISCUSSION R. J. PENTREATH: Is there any falling off in the stable zinc level in echinoderms with increasing distance from the shore? A.G. CAREY: It remains fairly constant. R. PENTREATH: Is the 65Zn present on the exoskeleton or in internal tissues? When 65Zn is accumulated from water only, it is adsorbed mainly on the exoskeleton.Is there perhaps a variation with the surface-to-volume ratios of asteroids? If the accumulation is largely in the gonads in echinoids, one would expect a large seasonal variation.Although water uptake experiments have shown that most of the 65Zn is accumulated on the exoskeleton, the concentration factor is still small.Your results would thus support our view that the main route of zinc intake is through the food chain. A.G. CAREY: The 65Zn is present in internal tissues; the levels in the test are very low or even undetectable.There is a seasonal change in the 65Zn content, caused by seasonal changes in the gonads.I would agree that the main route of radiozinc is through the food web. The surface-to- volume ratio would vary in different asteroid species, but I have not investigated the effect of this on 65Zn specific activities.I would suspect it to be very small. R. FUKAI: Can you say why there is a maximum specific activity of zinc at the mid-depth of 60 to 100 m? A.G. CAREY: The maximum is probably due to a combination of two factors, namely variation of transport rate with particle size and distance from the source. Near shore the coarser sand is probably relatively "older" than the silt at mid-depths.Still farther off shore the specific activity declines because of increasing distance from the Columbia River mouth. M. BERNHARD: Why do you express your data in terms of specific activity only. Would it not be pertinent to present stable zinc data? A.G. CAREY: The specific activity approach minimizes the variability caused by differences in the zinc concentrations in organisms. Data on stable zinc and 65Zn should have been reported here for the sake of completeness; they will be included when the paper is published in the proceedings. M.H. FELDMAN: While we are on the subject of data, are you able to suggest any reason for the fact that in Fig. 5 there is a considerably greater scatter in the data obtained off Tillamook than in those obtained off Newport? The higher specific activity specimens are noticeably more erratic than the lower. A.G. CAREY: The Tillamook Head transect is much closer to the river, and the 65Zn specific activities have a high variability because of pronounced seasonal changes in the position of the Columbia River plume 217

IAEA-SM -158/ 17 305 and hence in the availability of 65 Zn to the benthic fauna.I think this is the main cause of the large variations in the specific activities. Along the Newport transect, on the other hand, the variabilityhas been dampened owing to the greater distance from the Columbia River mouth. H.A. VANDERPLOEG: The reason for the higher variabilityin the 65 Zn specific activity values for the THB8 holothurians shown in Fig. 5 may in part be the higher specific activities of these animals. Under equilibrium conditions, differences in the zinc uptake rateamong holothurians will necessarily result in greater variabilityamong animals with higher specific activities, i.e. those eating food of higherspecific activity. This principle also applies to a period when 65Zn is decreasing in the animals and their food. 208

RLO-2227-T12-27 Reprintedfrom: InternationalAtomicEnergyAgency. 1973.Radioactive Contamination of the Marine Environment. [Proceedings of the Symposium on the Interaction of Radioactivecontaminants with the Constituents of the MarineEnvironment,SeattleWashington,10-14 July1972]. IAEA,Vienna. IAEA-SM-158/14

RADIOECOLOGY OF BENTHIC FISHES OFF OREGON

W.G. PEARCY, H, A, VANDERPLOEG Department of Oceanography, Oregon StateUniversity, Corvallis, Oregon, United States of America

Abstract

RADIOECOLOGY OF BENTHIC FISHES OFF OREGON. Gamma-emitting radionuclides were found in benthic fishes from depths of50-2800 in off the Oregon coast from 1964-1971; 65Zn60Co, "Mn,t"Ce, "'Cs and40K were present,Zinc-65, originating mainly from the nuclear reactorson the ColumbiaRiver, was the predominantartifically inducedradionuclide. Levels of65Zn per gram and specific activities of 65Zn decreasedmarkedlyin several speciesof fishes between 1965 and 1971because ofthe shutdown of reactors,This decrease was greater for small than for large Lyopsetta exilisduring1970-1971.Specific activities decreased with increasing depth, both for individual species inhabiting broad depth ranges and for different species inhabiting different depths.Specific activities of 65Zn were inversely related to body size for L. exilis and Sebastolobus. Other variations of 65Zn were related to trophic position.Fishes that preyed on low trophic level pelagic animals had higher specific activities than fishesthat preyedon benthic invertebrates.Such pelagic feeders may play an important role in accel- erating the transport of some radionuclides or elements to the sea floor.

1. INTRODUCTION

This paper examines the artificialradioactivity of benthicfishes off Oregon and, more specifically,how thespecificactivity of zinc-65 (nCi 65Zn/gtotalZn) varies among species, size of fishes,depthof capture, years and seasons, and feeding habits.Differences may be useful for identifyingpathways ofzinc accumulation from a surfaceinput and may provide information on the ecology of this important group of animals whichcomprises a significant portion of the biomass of benthic communities. Until recently the most important source of artificial radionuclides in the ocean off Oregon wasthe ColumbiaRiver [ 1 ] ( 2 ],Radionuclides were mainlyinduced byneutron-activation of elements in river water used to cool "single-pass", plutonium-production reactors at Hanford, Wash- ington.Most ofthe radionuclides produced had a short half-life and decayed to undetectable levels duringthe 370 mile tripdownthe Columbia River to thePacific Ocean.Starting in 1964 reactorsat Hanfordwere se- quentiallyshutdown, and in January 1971 the last reactor inducing radioactivityof coolant water was deactivated.This study encompasses the years when radioactivity introduced into the Pacific Ocean was declining.

The distribution of Columbia River water in the ocean depends largely on seasonal wind patterns and ocean circulation.During the summer, when winds are usually from the north, the Columbia River waters

245 2Oj

246 PEARCY and VANDERPLOEG

are pushed to the south off Oregon, often as a distinct plume of low- salinity water.During the winter, when southerly windsprevail, the river waters are found to the north largely along the coastof Washington [ 3]. Radioactivity introduced by the Columbia Riverhas beentraced in surface water by chromium-51 content [ 3 ] [ 4 ] andin sedimentsby zinc- 65 and cobalt-60 [3 ][ 5 Of the radionuclides introduced into the Pacific Ocean by the Colum- bia River, zinc-65 is the mostcommon in marineorganisms.65Zn, with a 245 day half-life,has been reported in many speciesof plankton,nekton and benthos off Oregon [61[7][$][91[101.

2. METHODS Fishes were collected with a 7-m semi-balloon shrimp trawlor a 3-m beam trawl on the continental shelf, slope and abyssal plain off Ore- gon from 1964-1971.

Fishes were sorted from the trawl samples and frozen in plastic bags at sea taking caution to prevent trace-metal contamination.In the laboiratory ashore they wereidentified,measured (standard length), weighed and dried to a constant weightat 65°C.To provide enough material forradioanalysis,several individuals of a species often constituted onesample.Afterdrying,samples were weighed, ashed at 450° C in a muffle furnace, ground to a fine powder with a mortar and pestle, and packed in 15cm3 plastic counting tubes for gamma-ray counting.

Samples were radioanalyzed in the well of a 12. 7 x 12. 7-cm sodium iodide (TI) crystal with a 512 channel pulse-height analyzer.Counting time was 100 or 400 minutes, depending on the radioactivity of thesample. Readout information was stored on paper punch tape, transferred to computer cards and analyzed by a least squares program to give radioactivity per gram ash, corrected for decay.

Zinc concentrations were determined on subsamples of ash using an atomic absorption spectrophometer (Perkin-Elmer Model303).These samples were first digested in concentrated HN03 and diluted with 0. 37N HC 1.

3. RESULTS Gamma-ray Spectra Zinc-65 was the predominant artificial gamma-emitting radionuclide in all benthic fishes.It produced a prominent photopeak and was present above minimum detectable activity' in virtually all samples.Besides 65Znand naturally-occuring 40K, other radionuclideswere occasionally

Minimum detectable activity, 3(N)1/2, where N=the background count, was 5 pCi for a 100-min count and 2 pCi for a 400-min count for bSZn, 210

IAEA-SM-158/14 247

FIG. 1.Gamma-rayspectrum of a 440-mm Merluccius productus (Pacifichake)collected at a depthof 311 m, March 1966.

evident above background.Distinct photopeaks of 54Mn and 137Cs were sometimes present in the spectra (Fig.1);they occurred most often in Anoploploma fimbria, Merluccius productus and Atheresthes stomias, all large carnivores. 6OCo and 144Ce were recorded rarely butno clear photopeaks were evident. 65Zn andSize of Fishes

Fish of thesame speciesbut of differentsizesometimes had values of 65Zn radioactivity per gram that were inversely related to body weight or length.Table I shows the amount of total zinc, 65Zn radioactivity and specific activity for different sizes of two species of Sebastolobus caught

TABLE I.ZINC, ZINC-65 RADIOACTIVITY AND SPECIFIC ACTIVITY FOR DIFFERENT SIZES OF Sebastolobus spp. CAUGHT AT THE 800-m DEPTH OF TILLAMOOK HEAD, OREGON, 13 JULY 1969

Average wet }1g Zn/g Number in wt. of fish wet pCibSZn/g Specific Sample in sample (g) weight wet weight Activity Species nCi/gZn

9.S. altivelis 42 5 10.4 0.161 15.4 avelisa tii velis 23 16 9.6 0.109 11.3 S. 10 48 8.6 0.082 9.5 s. ativelis 5 112 8.3 0.080 9.6 S. a love}' s 2 167 7.7 0.070 9.1 active is 2 249 5.4 0.053 9.7 S. alascanus 573 6.6 0.036 4.6 .as rascanus 761 6.6 0.027 4.0 a as aanus 983 6.3 0.026 4.1 248 PEARCY and VANDERPLOLG

19 0 70 8 16 I% 70 2 9 ID 70 3 10 % 70 3 12 IV 70 10 5 XI 70

4 10 Y 70 I I 8 %II 70 5 27 V 70 12 4 I 71 6 19 392 70 3 21 III 71 7 25 VIII 70 14 27 IV 71

I I I I I I 1 1 20 40 60 80 100 Weight (g)

FIG. 2.Regression lines of the 55Zn specific activities versus wet weight of Lyopsetta exilis for 7.4 collection dates in 1970 and 19'71, at one station.With order of ma nitude increase in weight of each spec- iesthe concentration of Zn andb5Znboth decreased.Specific activity of 65Zn also decreased significantly, indicating that the relative decrease in 65Zn was greater than that for Zn.The specific activities of Lyopsetta exilis, a common flounder, were also inversely correlated with body size. Fig. 2 shows the regression lines between size and specific activities for 14 different collection periods in 1970 and 1971.The slopes of all but two of these individual regression lines are significant (P < 0. 05).Therefore the effect of size must be considered when comparing spatial or temporal variations of specific activity of these species. Unlike Sebastolobus and L. exilis, the 65Zn specific activity of another flounder, Micros omus pacificus, was relatively independent of size.The slope of the linear regression between specific activity and weight was not significant for fish ranging between 96 and 846 g during April-May 1970 (P > 0. 5, n = 27), 42 and 318 g during June 1970 (P> 0. 2, n 11), and 200 and 1050 g during July-October 1970 (P> 0. 5,is = 40). To understand this inconsistency we must examine the reasons for expect- ing lowered specific activity with increased weight.

The model of Vanderploeg [ 11] helps to explain these specific activity-weight relationships.His model assumes zinc input is proportional to feeding intensity and thata ,the rate of zinc input per body burden of zinc, will decrease with weight, W, as will feeding intensity per unit weight.Using the respiratory and weight relationships of Win- berg [ 12 ] and the stable zinc- size relationship for M. pacificus, a decreases with W according to ati CW-0, 3 where C is a constant for a given temperature.It can be shown that in the steady state, this relationship implies that smaller fishes will have 212

IAEA-SM-158/14 249

a higher specific activity than larger fishes, assuming similar diets. From equation 9 given by Vanderploeg [ 11],the specific activity of a fish, S, under equilibrium conditions is given by a F S = + a where F = the equilibrium or constant specific activity of the prey, and A = 2.83 x 10'3, the physical decay constant for zinc-65.If, for example, a =1 x 10-3, a value obtained for M. pacificus [ 11],then S = 0. 26F.If a is doubled, then S = 0. 41F.Moreover, rapid growth, a concomitant of increased feeding intensity, implies increased a and thus higher specific activity.Since rapid growth characteristically occurs when fish are small relative to asymptotic size, the growth effect would tend to augment the size effect in a given species.

The conflicting results of the specific activity - weight regressions among the above species are clarified using the model.First, the relative size ranges of M. pacificus were smaller than those of the other two species. The sizes of Sebastolobus spp, and L. exilis ranged over more than two and one orders of magnitude respectively.In contrast, the size ranges of M. pacificuswere lessthan an order of magnitude. Because a is halved with each order of magnitude increase in weight, a specific activity - weight effect would not be strongly defined for M. pacificus,Second, these fishes were not in equilibrium with their food during this period of reactor shutdown.Under certain nonequilibrium conditions, the inverse relation between specific activity and size could diminish.

Long-Term Declines Short-term decline of radioactivity after temporary shutdown of Hanford reactors has been reported by Watson et al.[ 131 for freshwater animals and by Renfro and Osterberg [ 14] for 65Zn specific activities of the flounder Platichthysstellatusin the Columbia River estuary.Our study showslong-term decreasesof 65Zn inmarine fishesattributable to the reduced operationsof reactorsat theHanfordPlant on the Columbia River.Eight "single--pass" reactors were phased-out betweenDecember 1964 and January 1971.

The average specific activity of Microstomus pacificus decreased by an order of magnitude between 1965 and 1970 (Fig. 3), as the number of reactors decreased from five to one.Two other species of large flounders Glyptocephalus zachirus and Atheresthes stomias, showed similar decreases of specific activity over this same interval (Fig. 4). The specific activity of Lyopsetta exilis also decreased significantly during 1970-1971.This is shown by plotting the 65Zn specific activity for two sizegroups,5-20 mm and 45-60 mmfish,inFig.5A.Thesignifi- cant slope of the least squares fit to these points (P<0.001 and P<0. 01 for the small and large fish respectively) substantiates the decline of radioactivity associated with reactor shutdowns. 3

250 PEARCY and VANDERPLOEG

100

SPEC/F/C ACT/v/TY 60

C N 60

N G 40 C

6 REACTORS 4

1965 1 1 1966 1 1967 1 1968 1969 1 1970

FIG. 3.Specific activity of s5Zn for Microstomus pacific us collected off central Oregon, 180-270m, 1965 to 1971.The lower figure shows the number of Hanford nuclear reactors in operation during this period.

J13

1 u 0 G/yplocepho/us zachirus 140 Cl A/hereslhes s/omhs a \ 120 M hti 13 100 CV

60 O O tiv O 40 a

13 RY 20 0 23 W 0 I IV 1965 1966 1967 1968 1969 1970

FIG.4.Specific activity of stZn for two flounders, 1965-1970.Solid symbols denote fish <200 mm, half- solid symbols fish between 200 and 700 mm, andopen symbols fish >300 mm. 214

IAEA-SM-158/14 251

4 B.

0 4 ) T ) {

40 A.

30 V b 20

JIF A M I J J A ST0rN1 TF T A +M 1970 1971

FIG. 5A(below).Specific activities of 65Zn for L o etta exilis during 1970 and 1971.Each point represents a sample.Lines are least squares fits to the points. Upper regression line is for fish 5-20g wet weight; lower line for fish 45-60 g wetweight. B,(above)The slope of the specific activity versus weight relationships of Fig. 2 plotted for different collections during 1970 and 1971.Slopes and their standard errors are shown. For the least squaresfit,each point was weighted by the inverse of itsstandard error.

Fig. 2 shows that both the slopes and the Y- intercepts of the regression lines for L L. exilis usually decreased with time.The regression line of these slopes vs, time is significant (P<0. 05; Fig. 5B). These trends, and the significantly (P<0. 01) larger slope of the regression line for small than for large fish in Fig. 5A, indicate thatsmall L. exilis responded more rapidly to decreased levels of 65Zn in the environment than large L. exilis.The Zn turnover of small fishwas fasterand/or the specific activity of their prey decreasedmore rapidly.

SeasonalVariations

Although seasonal variations are notapparent in the data for L. exilis from 1970 and 1971 in Fig, 5, 22 analyses ofearlier collections, when 65Zn radioactivity levels were higher,demonstrate that specific activities were indeed higher during May-September,the time of year when the plume is usually located off Oregon, thanduring October- April, when most Columbia River flows north alongthe Washington coast (Table II). 21

252 PEARCY and VANDERPLOEG

TABLE II.ZINC AND ZINC-65 IN Lyopsetta exilisCOLLECTED FROM JUNE 1964 TO JANUARY 1967, 200m DEPTH OFF NEWPORT, OREGON All fish were between 100 and 200mm in length

May-September October-April P

pCi b5Zn/g ash-free dry wt. 12.6 8.21 >0.1

ug total Zn/g ash-free dry wt. 63.9 72.1 >0.3

nCi65Zn/g Zn 224 110 <0.02*

Depth of Capture

The average specific activities of benthic fishesare listed according to depth of capture in Table III for 1964-1967,the period before rapid decreases of specific activities (Figs. 3 and4).Specific activities generally decreased with depth.This trend is evident despite the large variability listed for individual species.Variation is due to aforementioned factors and those cited by Renfro and Osterberg [14]

Specific activities of deep-sea macrouridson the abyssal plain (2700-2900m) are between one andtwo orders of magnitude lower than values for fishes from the upper continental shelf (50-100m).This general trend for decreasing specific activities withdepth is believed to be mainly caused by the increased time required for 65Zn vertical transport of from surface waters into the deep-sea.The difference in average specific activities between fisheson the upper shelf and the abyssal plain is equivalent to 4. 3 half-lives of 65Znor about 3 years.Larger 65Zn input into surface waters on the shelf isa possibility but the geographic location of the Columbia River plume basedon salinity and radioisotopes [ 3][ 41[15] and echinoids [ 9] all show that the plume commonly is found intermediate distances off Central Oregon during thesummer and usually is not confined to shallow water over the inner shelf whereupwelling is prominent.

Specific activities of individual species capturedover a wide depth range also decreased with increased depth of water (Fig. 6).Microstomus pacificus had highest specific activities in shallowwater (less than 200m depth) and values were lower by 10 times in deepwater (800 m).Note that larger fish were usually captured in deepwater, a factor that could confound interpretation of specific activitieswere it not for the relative independence of specific activities and size for large fishof this species over moderate size ranges.The 65Zn specific activities of Sebastolobus spp, also decreased in deeper water during 1965-1967 but not in 1970when only one reactor was in operation.Data on hagfishes (Eptatretus spp. ) suggest a change of specific activity with depth in 1966 butnot in 1970 when all had low specific activities. 216

IAEA-SM-158/14 253

o I - Sail t96S X 1966 - 11967 O 11-1Q 1970

Sebostbus

20 0 O 0 _ to I I 1 0 400 800 1200 1600

so Eptatretus 0 a-syI -3M 570 566 40

20 0 0 0 _0l_.__ I 1 I I 0 400 800 1200 1600 Depth (m)

FIG. 6.Specific activity of 65Zn at different depths for several benthic fishes.Symbols indicate different

collection times and size of fishes.The symbols for Microstomus acp ificus:solid <200 mm, half-solid 200-300 mm, open >300 mm; for Sebastolobus spp.: solid <100 mm, half-solid 100-200 mm, open >200 mm;

for E tap tretus spp. all individuals were between 310-510 mm.

Even fishes from deep water had specific activities that varied with depth. A comparison of specific activities of Macrouridae collected in 1965 and 1966 from 1250-1600 m on the continental slope and from 2800m on the abyssal plain showed that the fishes from slope waters had highest specific :activities (P < 0. 05, n = 19). TABLE III.SPECIFIC ACTIVITY OF ZINC-65 IN SPECIES OF BENTHIC FISHES COLLECTEDON THE CONTINENTAL SHELF AND SLOPE AND THE ABYSSAL PLAIN OFFOREGON, 1964-1967

Specific Activity Depth Species Length No. of nCi65Zn/gZn (m) Range Analyses mean std. dev.

50- 100 Citharichthys sordidus 125-210 4 560 90

Cymatogaster aggregata 85-110 3 370 280 Eopsetta jordani 115-235 4 220 120

Isopsetta isolepis 135-250 4 80 10 parophrys vetulus 190-300 7 130 60 100- 200 Microgadus proximus 150-190 3 170 120 Sebastes spp. 160-250 6 140 90 Atheresthes stomias 260-380 4 170 110 Anoplopoma finbria 160-450 7 90 60 Xenopyxis latifrons 90-160 6 120 20 Lyopsetta exilis 70-200 33 140 110 Sebastolobus altivelis 125-200 6 100 20

Sebastes elongatus 170-250 5 80 20 Microstouus pacificus 245-360 14 70 60 Olvptocephalus zachirus 225-285 16 70 40 200- 400 Merluccius productus 420-530 3 50 30 Eptatretus spp. 300-460 8 30 10 400- 800 Sebastolobus alascanus 55-245 14 40 20 Anoplopoma fimbria 400-470 5 30 20 Mlcrostomus pacificus 130-360 9 30 30 800-1300 Macrouridae 260-250 5 30 30

1600-2300 Antimora rostrata 310-470 4 20 20 2700-2900 Macrouridae 200-570 25 10 10 218

IAEA-SM-158/14 255

TABLE IV. FEEDING HABITS AND 65Zn SPECIFIC ACTIVITIES OF PLEURONECTIFORMES (FLATFISHES) COLLECTED AT TWO DEPTH RANGES OFF OREGON, 1964-1967

nCi 65Zn/gZn Specific Activity 50-100m Food Habits

Citharichthys sordidus 560 euphausiids, shrimps, amphipods crab larvae Eopsetta 'ordani 220 shrimps, pelagic fishes, euphausiids

Isopsetta isolepis 80 gastropods, polychaetes, pelecypods

Parophrys vetulus 130 polychaetes, amphipods, pelecypods

100-200m

Atheresthes stomias 170 fishes, shrimps, euphausiids

Lyopsettaexilis 140 euphausiids,shrimps

Microstomus pacificus 70 polychaetes, ophiuroids, pelecypods

Glypotocephalus zachirus 70 polychaetes,amphipods

Food Habits Within the depth intervals listed in Table III, specific activities of some fishes are clearly related to feeding habits.Stomach contents of Pleu:ronectiformes(flatfishes)were analyzed and specieswithhighest specific activities were found to feed largely on pelagic prey such as euphausiids, pandalid shrimp, and fishes, whereas species with low specific activity fed on benthic prey such as polychaetes, ophiuroids, mollusks and amphipods (Table IV).This trend is also illustrated in Fig. 7 which shows the 65Zn specific activities of several species from one collection. Specific activities were highest forL. exilis andA. stomias, pelagic feeders, and lowest for M. pacificus and G, zachiruq benthic feeders.The 65Znspecific activities of the stomach contents of L. exilis and A. stomias from this collection were 40 nCi/g;they were 22 nCi g for M. pacificusl. x;13

256 PEARCY and VANDERPLOEG

Lyopse//o ex//is + + Atheres/hes s/om/os Glyplocepha/us zochirus O M/crostomus pocif/cus O £plo/re/us

0 0 C0 0 0 0 O 0

I I Lam 01 I I 1 I 0 200 400 600 800 1000 Weight (g)

FIG. 7.Specific activity versus wet weight of five species of benthic fishes all collected at one station on 10 May 1970.All weights less than 100 g are averages of more than one individual per sample.

In all instances specific activities of the prey were considerably higher than the fishes themselves.This decrease in specific activity follows from radioactive decay in passage up the food chain [16].

Based on this relationship between 65Zn specific activities and food habits we thought that Isopsettaisolepis, a flounder with a low 65Zn specific activity (Table III) and for which we could find no information on feeding habits, would be a benthic feeder.Subsequent examination of stomachs proved this to be the case (Table IV).Thus specific activity, in this instance, was used to predict feeding habits. Apparently some species of fishes captured in bottom trawls are more closely linked with the pelagic than the benthic food web. Some species are pelagic and reside at times just above the bottom. Some undertake diel vertical migrations swimming off the bottom at night, as reported for other species [ 171Others may reside on the sea floor but feed on pelagic animals that migrate close to the bottom during the day. Pandalid shrimp and euphausiids are vertical migrants [18][19] that are important food for several benthic fishes off Oregon (Table IV) and for Atheresthes stomias and Merluccius productus from northern California [20] . Pelagic feeders have higher 65Zn specific activities than benthic feeders because less time is evidently required for atoms of 65Zn to be passed to fishes through the pelagic food chain than through the benthic food chain and relatively less physical decays occurs.This lag explains why surface-living or predatory benthic invertebrates have higher 05Zn content than deposit-feeding infauna [6] [9] .The fact that euphausiids have relatively high concentrations of 65Zn compared to other animals [7] [Pearcy, unpubl. ] also supports this contention.Therefore pelagic- 220

IA EA-SM -158/ 14 257

feeding benthic animals may accelerate the transport of elements and energy from surface waters to the sea floor just as vertically, migrating organisms may accelerate transport into the deep sea [ 8] . The relationship between specific activity and weight for Lyopsetta exilis shown in Figs. 2 and 5 may in part be owing to the feeding habits of different size classes of this species [ 21]. The size,of ingested organisms was positively correlated with size of fish examined during some months. The small prey consisted of small pelagic crustaceans such as euphausiids and decapod larvae.The large prey were shrimps and fishes that often had lower specific activities than the small crustaceans they consumed [18] [21].Assuming L. exilis accumulates zinc through the food chain, small fish would be expected to have the highest specific activities.Diet may also be responsible, in part, for the more rapid decline of specific activity in the smaller L. exilis (Fig. 2).Possibly small pelagic crustaceans respond more quickly to decreasing environmental radioactivity causing, in turn, a more rapiddecreasein smaller L. exilis.

4. SUMMARY AND CONCLUSIONS

1. Although otherradionuclides werepresent, 65Zn was themostconspic- uous gamma emitter found inbenthicfishes off Oregon. 2. Therelationship between specific activitiesof 65Znand sizeof individuals was investigated for three species. A significant inverse relationship was found for two of these species. 3. Long-term decreases of 65Zn specific activities, illustrated for several species, were correlated with the shutdown of nuclear reactors on the Columbia River between 1964 and 1971. 4. Both the slope and the Y-intercept of the regressions of 65Zn specific activity on Lyopsetta exilis weight decreased with time during 1970 and 1971. This indicates that small L. exilis responded more rapidly to the decline of 65Zn in the environment than large L. exilis,The fast- er decrease of zinc-65 in smallfishis related to their size and the specificactivity of their prey. 5. Seasonal variations of 65Zn specific activities in L. exilis were observed that correlatedwithknown seasonal changes in the location of the Columbia River plume. 6. Specific activities of 65Zn decreased with increasing depth and were at least 10 times lower in fishes captured at depths of 2800m than 100m. Individuals of species collected over a wide depthrangedisplayed similar trends. 7., The fact that benthic fishes feeding on pelagic prey generally had higher65Zn specific activities than fishes feeding on benthic invertebrates suggests that these pelagic feeders are more closely coupled withthe source of 65Zn and accelerate its transport to thesea floor. 22 1

258 PEARCY and VANDERPLOEG

ACKNOWLEDGMENTS

We thank R. J. Eagle, H. Wyandt, I. L. Larsen andG. Wagner for their valuable assistance at seaand insamplepreparation and analyses, D. Hancock for identificationof organismsfrom fish stomachs, and R. Holton and N. Cutshall for reviewing the manuscript.We arealsoindebt- ed to A. G. Carey, Jr. who collectedsome of the fishes that we analyzed. We are grateful to the U. S. Atomic Energy commission (ContractNo. AT(45-1) - 2227, Task Agreement 12 for financialsupport (RLO-2227-T12- 2 7).

REFERENCES

[1] OSTERBERG, C. L., CUTSHALL, N., JOHNSON, V., CRONIN, J. JENNINGS, D. and FREDERICK, L., Some non-biologicalaspects of Columbia River radioactivity, Disposal of RadioactiveWastes into Seas, Oceans and Surface Waters (Proc. Conf.Vienna, 1966), IAEA, Vienna (1966) 321. [2] PERKINS, R. W., NELSON, J. L., HAUSHILD, W.L., Behavior and transport of radionuclides in the Columbia River betweenHan- ford and Vancouver, Washington, Limnol. Oceanogr. 11 2 (1966) 235.

[3] BARNES, C. A., GROSS,M. G., Distribution at sea of Columbia River water and its load ofradionuclides,Disposal of Radioactive Wastes intoSeas,Oceans and Surface Waters(Proc. Conf. Vienna, 1966), IAEA, Vienna(1966) 291. [4] OSTERBERG, C. L., CUTSHALL, N. CRONIN, J., Chromium-51 as a radioactive tracerof ColumbiaRiver water atsea,Science 150 (1965) 1585.

[5] CUTSHALL, N., RENFRO, W. C., EVANS, D. W., JOHNSON,V. G., Zinc-65 in Oregon- Washingtoncontinental shelf sediments, Third Radioecology Symp. (Proc. Conf. Oak Ridge,in press). [6] CAREY, A. G. Jr., PEARCY, W. G., OSTERBERG, C. L., Arti- ficial radionuclidesin marine organisms in the Northeast Pacific Ocean off Oregon, Disposal of RadioactiveWastes into Seas, Oceans and Surface Waters (Proc. Conf. Vienna, 1966),IAEA, Vienna (1966) 303. [7] OSTERBERG, C. L., PEARCY, W. G., CURL, H. C. JR.,Radio- activity and its relationship to oceanic food chains, J. Mar, Res. 22 1 (1964) 2.

[8] PEARCY, W. G., OSTERBERG, C. L., Depth,diel, seasonal, and geographic variations in zinc-65 or midwater animals of Oregon, Int. J.Oceanol, andLimnol, 12(1967)103. 222

IAEA-SM-158/14 259

[9] CAREY, A. G. JR., Zinc-65 in echinoderms and sediments in the marine environmentoff Oregon, Second Nat'l Symp. Radioecology (Pros, Conf. Ann Arbor, 1969) 380. [10] SEYMOUR, A. H., LEWIS, G. B., Radionuclides of Columbia River origin in marine organisms, sediments, and water collected from the coastal and offshore waters of Washington and Oregon, 1961-1963, U. S. Atomic Energy Comm. UWFL-86(1964). [11] VANDERPLOEG, H. A., Therate of zinc uptake by Dover Sole in the Northeast Pacific Ocean:Preliminary Model and Analysis. Third Radioecology Symp. (Proc. Conf. Oak Ridge,in press). [12]WINBERG, G. G., Rate of metabolism and food requirements of fishes, Belorusskovo Nauchnye Trudy Belorusskovo Gosudarstven- novo Univ. V. I.Lenina, Minsk (1956) [Trans. Fish. Res. Bd. Canada, No. 194].

[13] WATSON, D. G., CUSHING, C. E., COUTANT, C. C., TEMPLE- TON, W. L., Effect of Hanford reactor shutdownon Columbia River biota, Second Nat'l Symp. on Radioecology(Proc. Symp. Ann Arbor, 1969) 291.

[14] RENFRO, W. C., OSTERBERG, C. L., Radiozincdecline in starry flounders after temporary shutdown of Hanford reactors, Second Nat'l Symp,on Radioecology(Proc. Symp. Ann Arbor (1969) 372. [15] OSTERBERG, C. L., PATTULLO, J., PEARCY, W. G., Zinc-65 in euphausiids as relatedto ColumbiaRiver water off the Oregon coast,Limnol.Oceanogr.9 2(1964) 249.

[16]FOSTER, R. F., Radioactivetracingof themovement of an essential element through an aquaticcommunity withspecial reference to radiophosphorous, Publ. Staz.Zoologica Napoli 31 (Supplements) 34. [17] BEAMISH, F. W. H., Verticalmigrationby demersal fish in the Northwest Atlantic, J. Fish. Res, Bd. Canada 23 1(1966) 109.

[18] PEARCY, W. G., Verticalmigration of oceanshrimp,Pandalus jordani:a feeding and dispersal mechanism, Calif.Fish and Game 56 2(1970) 125.

[19]BRINTON, E., Verticalmigrations and avoidancecapability of euphausiidsin the California Current,Limnol. Oceanogr. 12 2(1967) 451.

[20] GOTSHALL, D. W., Stomach contentsof Pacifichake and arrow- tooth flounder from northern California, Calif. Fish and Game 55 1(1969) 75. [21] VANDERPLOEG, H. A., Dynamicsof Zinc-65 in benthic fishes and their prey off Oregon,Ph. D.Thesis, Oregon StateUniv. Lib. (1972), 223

260 PEARCYand VANDERPLOEG DISCUSSION B. PATEL: Do you have any data on the distribution of 137Cs and 144Ce in fish of various sizes? We have much data for nuclides such as 60Co, 65 Zn and 54Mn, but not for nuclides of less obvious physiological importance. W.G. PEARCY: Unfortunately, the number of samples containing quantities of 137Cs and 144Ce in excess of the minimum detectable activity was too small to permit such comparisons. D.M. MONTGOMERY: With regard to the decrease in the specific activity of 65Zn with increasing depth for a given species of fish, how can one be sure that a fish captured at a given depth is representativeof that depth? W. G. PEARCY: Movements of fishes from one depth to another may be a problem and probably contribute to the large variation of specific activity encountered in individual species (Table III).Seasonal bathymetric migrations of several types of flat-fish are known to occur off Oregon. M. BERNHARD: Can you say something more about the stable zinc concentrations? Are they constant, for example? W. G. PEARCY: The stable zinc concentration varies within species but normally not much. We plan to investigate the causes of this variation. D.R. YOUNG: Do you have any reservations about using the specificactivity approach in view of the possibility that equilibrium between Columbia River 65Zn and stable zinc in the marine ecosystem, possibly in morethan one physical and chemical state, may not have beenreached at the time of collection? Are you able to draw any conclusions regarding the extent to which equilibrium has been reached in the various samples analysed? W.G. PEARCY: Even if 65Zn is in the same form as stable Zn, organisms do accumulate some of the 65Zn from the Columbia Riverplume before it mixes very much with the seawater. We are looking into the possibility that organisms in the food chain accumulate 65Zn in a form different from that of stable zinc in seawater. W.R. VERMEERE: Were the animals and fish that formed the basis for the specific activity/depth diagrams collected from the same station, i. e. from the same vertical collection profile? W. G. PEARCY: All our samples were obtained by dragging trawls along the sea bottom.Stations were located at increasing depths along lines perpendicular to the coast. E.A. PRYZINA: Would you care to comment on where the65Zn is localized in the fish you studied? W. G. PEAR CY: The 65Zn concentration in relation to wet weight (pCi/g) was usually highest in the gastrointestinal tract (with contents), liver and flesh, in that order.All the measurement results reported in this paper are based on the analysis of whole fish minus the contentsof the stomach and intestine. R. J. PENTREATH: Although the zinc concentration in the liver ishigh, it represents only a small part of the total amount of zinc present inthe fish.Data that we have collected indicate that the content of the muscle, bone and skin accounts for most of the total on a whole-body basis. W. G. PEARCY: In my reply to Mr. Pryzina I was in fact speaking of concentration rather than absolute content.If my memory serves me correctly, most of the 65Zn in our fish is located in the muscle and bone. 224

IAEA-SM-158/14 261

D. E. ROBERTSON: If one is analysing whole organisms from the benthie and upper -water- column communities, how valid are the inter- comparisons of specific activities when the benthic organisms contain large quantities of sediment in their digestive tracts? W.G. PEARCY: This does not appear to be important. The zinc concentrations are very highin these animals, and often the quantity of sediment present is not very large.Moreover, the specific activity of the sediment does not differ greatly from that of the fauna.It should also be pointed out that the fish accumulate Zn and 65Zn from the stomach contents of their prey as well as from the prey organisms themselves. 225

RLO-2227-T12-28 Reprinted from: InternationalAtomic Energy Agency.1973.Radioactive Contamination of the MarineEnvironment[Proceedings of the Symposiumon the Interaction of Radioactive Contaminants with the Constituentsof the Marine Environment, Seattle, Washington, 10-14 July, 1972] IAEA, Vienna. IAEA-SM-158/8 EFFECTS OF OCEAN WATER ON THE SOLUBLE-SUSPENDED DISTRIBUTION OF COLUMBIA RIVER RADIONUCLIDES*

D.W. EVANS, N.H. CUTSHALL School of Oceanography, Oregon State University, Corvallis, Oregon, United States of America

Abstract

EFFECTS OF OCEAN WATER ON THE SOLUBLE-SUSPENDED DISTRIBUTION OF COLUMBIA RIVER RADIO- NUCLIDES. The relationships of dissolved concentrations of Hanford radionuclides with salinity in the Columbia River estuary were interpreted in terms of the exchange of the radionuclides between dissolved and suspended particulate phases. Both 65Zn and 54Mn carried bythe ColumbiaRiver were partially desorbed from suspended particulate matter upon mixing with ocean water in the estuary.Experiments in which ocean water was added to Columbia River water and to suspended particulate matter collected on filters confirmed the partial desorption of 65Zn and 54Mn from the particulate phase. The percentage of 65Zn and 54Mn desorbed varied with experimental approach but desorption of 65Zn seemed to lie in the range 15 - 45% and "Mn in the range 30 - 6001o. None of the experiments revealed any effect of salinity upon the soluble-suspended particle distribution of "Cr, 124Sb or 46Se.Dissolved concentrations of these nuclides varied inversely with salinity.There was no evidence that any of the radionuclides studied was removed from solution by flocculation, precipitation or adsorption as the result of mixing with ocean water. Ocean water contact partially removed 54Mn but not 65Zn, 46Scor 60Co from Columbia River bottom sediments transferred to the marine environment. The inability of ocean water to desorb 65Zn from bottom sediment contrasts with its action with suspended particulate 65Zn.

1. INTRODUCTION

The distribution of radionuclides between dissolved and particulate phases is important in determining their modes of physical transport and biological uptake in aquatic environments.Changes in environmental conditions can alter this distribution.For radionuclides introduced into rivers, the most extreme change in conditions probably takes place where rivers enter the ocean.In estuaries major redistributions of radionuclides might occur because of gradients in salinity and chemical composition, and possibly in temperature and biological activity as well. Laboratory studies with radioactive tracers adsorbed on a variety of artificial and natural substrates have simulated estuarine mixing.Fukai [ 1 ] and Kharkar et al.[ 2] reported partial desorption of several radionuclides from various substrates upon contact with seawater. Studies of the behaviour of trace metals under more natural conditions have allowed inference of the behaviour of their analogous radioisotopes. De Groot et al, [3] reported the "solubilization" from fluvial sediments

*This work was supported by the USA EC contract AT(46-1) 2227, Task Agreement 12. (RLO-2227-T12-28).

125 22.6

126 EVANS andCUTSHALL

0025 50 Mm Oregon

FIG. 1. The Columbia River from Hanford to the Pacific Ocean.

of a number of heavy metals in the Rhine and Ems estuaries.Conversely, Lowman et al. [4] have noted the flocculation and precipitation of several trace elements dissolved in river water after mixing with seawater. Conclusions about the behaviour of radionuclides in estuaries appear to depend not only on the properties of the radionuclides of interest, but also on the experimental approach. We report resultsof some experiments aimed at detecting the exchange of some Columbia River radionuclides between suspended particulate matter and solution consequent on mixing with seawater. The Columbia River is unique in having had, until recently, a nearly continuous input of radioactivesubstancesat relatively high activity levels. These radionuclides were formed by neutron activation of corrosion products and dissolved trace elements in the coolant waters of Hanford plutonium production reactors. These radionuclides entered the river largely in dissolved form in coolant water [ 5] .During downstream flow, many radionuclides became increasinglyassociatedwith suspended matter and bottom sediments.At Vancouver, Washington, 425 kmdownstream (See Fig. 1), more than 80%of 85Zn, 60 Co, 46 Sc, and 54Mn could be removed from river water by filtration.The radionuclides 51Cr and 124Sbremained largely in dissolved form (greater than 90%) although significantfractions had been deposited with bottom sedimentsalongthe intervening stretch of the river [ 5 ] . Johnson et al.[ 61 leached bottom sediments collected from the Columbia River with seawater.Between 16 and 74% of the associated 54 Mn was eluted during the nominal one-hour contact periods.Less than 3% of the'65Zn and negligiblepercentagesof the 51Cr and 46Sb were removed. To the degree thatbottom sediments represent suspended sediments, these studies simulated the behaviour of radionuclidesassociatedwith suspended matter in the Columbia River estuary.The possibility of adsorptive uptake ,by sedimentswas not examined. 227

IAEA-SM-158/8 127

2. RADIONUCLIDE-SALINITY RELATIONS IN THE COLUMBIA RIVER ESTUARY Dissolved and suspended particulate gamma-emitting radionuclides "Cr, 65 Zn, 46 Sc, and 124Sbwere sampled at locations of different salinities in the Columbia River estuary.Surface water was collected at mid-channel during a three-hour period, 30 July, 1970, by centrifugal pump on board R/V SACAJAWEA (see Fig. 2 for locations).Duplicate 50-litre filtrate samples were obtained by pressure filtering 100 litres of water through a single 265-mm-diam. 0.45-µm pore size, membrane filter (Gelman Type GA-6) sandwiched between two glass-fiber prefilters (Gelrnan Type E) held in a specially constructed polyvinyl chloride support structure. The filtrates were contained in 8-mil polyethylene bags supported by 70-litre plastic barrels. Water samples for salinity determination were collected simultaneously and analysed in the laboratory using an inductive salinometer. Coprecipitation with Fe(OH)3 was used to concentrate the dissolved radionuclides from the filtrate.Ten millilitres of carrier solution (0. 193M K2Cr2O7, 0. 306M ZnSO4 and 0. 368M MnSO4) were added and mixed with the filtered water.Thirty minutes was allowed for isotopic equili- bration, and 25 ml 0. 7M FeSO4 was added and mixed to reduce 51Cr and stable chromium from the +6 to the more readily absorbed +3 oxidation stage [ 71.After a second 30-min period, 25 ml 2M FeCl3 was added and mixed.After a final 30-min period, 25 ml of concentrated NH4OH was added and mixed to raise the pH, precipitate Fe(OH)3, and to co-precipitate the radionuclides and stable carriers.One hundred millilitres of 0. 05% Dow Separan NP10 was added and mixed to promote flocculation and settling of the precipitate.The supernatant was siphoned off, and the precipitate slurry drained of water by filtration and dried. The dry precipitate was ground to a powder, and a 1% sub-sample weighed out and dissolved in 0. 36N HC1 for analysis of Zn and Cr by atomic absorption spectrophotometry. The efficiency of recovery of 51Cr and 65Zn was estimated from the recovery of the stable carriers, since the amount of Cr and Zn added was known. Chromium recoveries ranged from 76 - 95% with mean 87%.Zinc recoveries ranged from 75 - 91% with mean 83%.

Oregon 4,, FIG. 2. The Columbia Riverestuaryshowingsamplinglocations. 30 July. 1970. 228

128 EVANS andCU!ISHALL

50r 0.5r 51Cr 65Zn

40 0.4

30. 0.3

20 0.2

10 0.1

1 i 1 0 20 400 20 40

0.4 0.4 124Sb 46SC

0.6 0.3

0.4 0.2

0.2 0.1

t I 0 20 40 0 20 40 Salinity ('.bob

FIG. 3.Concentrations of dissolved radionuclides in the Columbia River estuary, 30 July, 1970.Solid lines represent least squares fit to all data points. Dashed line is least squares fit to data points at salinities 4.98%o and above only.

The remaining 99% of the precipitate was packed into a 12-m1 plastic counting tube and radioanalysed using a 5 in. X 5 in. Nal(Tl) well-crystal coupled to a Nuclear Data 512-channel analyser.The output was a composite spectrum of the individual gamma spectra of the radionuclides present. After subtraction of background, the activities of the individual radionuclides were determined by a least squares fit of the individual spectra of radionuclide standards.Corrections for decay were made and the standard error of estimate of the activity of each radionuclide was calculated. 'Filters and suspended matter were digested for 3 h in 200 ml of a 1:1 mixture of concentrated HCl and concentrated HNO3 at 100°C.After cooling, the liquid was filtered from the solid residuum. The residuum was washed three times with 0. 36N HC1 and the washings added to the filtered liquid.This mixture was then evaporated to dryness under a heat lamp. The resultant dry froth was ground to a powder and packed in a counting tube for radioanalysis, effecting a significant reduction in volume with greater than 95% recovery of the radionuclides of interest. The dissolved concentrations of 51Cr, 65Zn, 46sc, and 124Sb have been plotted versus salinity in Fig. 3.Corrections for recovery efficiency have 229

IAEA-SM-158/8 129

5 0.5- s1Cr 65Zn 4 0.4

3 0.3

2 0.2

0.1

I I I I 0 20 40 0 20 40 0.6 124Sb 46SC 0.5

0.4

0.3 All samples less than 004pCi/L 0.2

0.1

I I I I I I 1 0 20 40 0 20 40 So/in/y (%e)

FIG. 4. Concentrations of suspended particulate radionuclides in the Columbia River estuary. 30 July, 1970.

been made for 51Cr and 6SZn.Total recovery has been assumed for 46Sc and 1245b, Particulate radionuclide concentrations are plotted in Fig. 4. If dilution by seawater were the only cause of changes in radionuclide concentrations in the estuary, a plot of dissolved concentration versus salinity should be linear for each radionuclide.The dissolved concentrations of the radionuclides of interest are negligibly small in the undiluted seawater mixing in the estuary; therefore, such a linear plot should havea salinity intercept equal to the salinity of the mixing seawater.This salinity should be between 32. 6 and 33. 4%[ 8 1[ 9 ]. The least squares salinity intercept and linear correlation coefficients are shown in Table I. The concentrations of 124Sb, 46Sc, and 51Cr have significant linear relationships with salinity at the 95% confidence level as determined by the F test using the duplicate data values to estimate pure error [ 10].Both 124Sb and 46Sc havesalinity intercepts very close to that of the mixing seawater.That of 51Cr is unexpectedly high, possibly indicating changing input concentrations from the river during the mixing period of the waters in the estuary.It could not be due to desorption from suspended particles because the concentrations of this possible source are too low to account for the increase. 230

130 EVANS and CUTSHALL

TABLE I.LEAST SQUARES FIT OF DISSOLVED RADIONUCLIDE CONCENTRATIONS TO THE LINEAR DILUTION MODEL, USING ALL DATA POINTS

Linear correlation Radionuclide Salinity intercept (%o) coefficient

51Cr 38.5 0.965 65Zn 38.2 0.860 124Sb 32.8 0.969 46Sc 32.5 0.867

In contrast, 65Zn does not fit the linear model.Dissolved 65Zn concentration reaches a maximum at salinity, 4. 98%, rather than at 0%0.At salinities above 4. 98%0, the dissolved 65Zn concentration:salinity relation is highly linear, with a correlation coefficient of 0. 971 and a salinity intercept of 32. 6%0, suggesting that only dilution by seawater acts to change dissolved 65Zn concentrations in this salinity range. The dissolved 65Zn concentrations at salinity 1. 12%0 and 2. 75%o are significantly less than that predicted by the linear regression curve through the data points at salinity 4.98%, and above.The difference between observed and predicted values can be explained if dissolved 65Zn was added between salinity 1. 12 and 4. 98%0. The only source of this added 65Zn must be 65Zn associated with particulate phases.At salinity 1. 12%, the difference that must have been added is about 0.15 pCi/litre.The suspended 65Zn concentration at salinity 1. 12%, was 0.46 pCi/litre.Therefore, about 33% (= 0. 15 pCi/litre divided by 0.46 pCi/litre times 100%) of the suspended 65Zn would have been desorbed. Nothing can be said about possible desorption at salinities less than 1. 12%0. The linear dilution model cannot be applied in the same way to the suspended particulate radionuclide concentrations because settling out or resuspension of particulate phases in the water column can alter the concentration of suspended radionuclides without requiring exchange with the dissolved phase. Data collected b y Hanson [ 11 ] show more dramatically the contrasting behaviour of 51Cr and 65Zn.Water samples were collected during a 25-h period, 21-22 May, 1966, along a vertical profile from the Point Adams Coast Guard Station pier (see Fig. 2 for location).Samples were processed and analysed much as described above with the exception that recovery ef- ficiencies were not determined and are assumed to be 100%.The radionuclide "Sc was relatively less abundant at this time, making possible the determination of 54Mn, with which it interfered.The original data reduction did not include 124Sb; rather, it was later determined from the residuals.Con- sequently 124Sb is reported in cpm rather than pCi because counting efficiency was not known.Dissolved radionuclide concentrations are plotted versus salinity in Fig. 5.Both 51Cr and 124Sb show strong linear relations with salinity, with correlation coefficients of 0. 936 and 0. 830 respectively. The salinity intercepts were 31. 7 and 38.4%0, respectively.The latter is higher than expected largely as a result of two outlying high salinity data points.These radionuclides appear to be conservative in the estuary, exhibiting no net exchange between soluble and particulate phases. 231

IA EA -S M-158/8 131

800 8 65Zn S1Cr

600 6 I~t

400 4[

200 2

I i I I I 1 I i 0 20 40 0 20 40 0.8 r *124Sb 54Mn 6 ., L 0.6

4 0.4

2L' 0.2 1

1 1 0 20 40 0 20 40 Salinity (%ol

FIG. 5. Concentrations of dissolved radionuclides at Point Adams Coast Guard Station, 21-22 May, 1966. *124Sb is reported in units of cpm/litre.

Neither 65Zn nor 54Mn shows a strong linear relation with salinity. They appear to be non- conservative; their concentrations at salinities above 5%o are greater than would be predicted by simple dilution with seawater. An estimate of how much greater can be obtained by plotting the concentrations on a seawater free basis versus salinity.Measured concentrations were divided by the fraction of fresh (Columbia River) water present to obtain these adjusted concentrations (fraction freshwater equals 1 - S/33. 0 where S is the measured salinity and 33. 0 is the salinity of the mixing seawater).Dissolved concentrations plotted on this basis are shown in Fig. 6. There does not appear to be a minimum salinity above which only dilution takes place. It appears that 51Cr and 124Sb are constant with salinity, whereas 65Zn and 54Mn tend to increase with increasing salinity.This increase results from addition of dissolved 65Zn and 54Mn, probably by desorption from suspended or bottom sediments. Using the mean of the adjusted concentrations of the five samples with salinities less than 1%o as the best estimate of the dissolved concentrations of inflowing Columbia River water, the concentration (on a seawater free basis) added by desorption at each salinity sampled can be calculated.The mean of these five samples is 2. 6 pCi/litre 232

132 EVANS and CUTSHALL

800 65Zn

6 12

400 8

20 40 0 20 40 1.6

1.2 6

0.8 4 ,.

2 0.4

0 20 400 20 40 Salinity (%o)

FIG. 6.Concentrations of dissolved radionuclides, on a seawater free basis, at Point Adams Coast Guard Station, 21-22 May, 1966. *124Sb is reported in units of cpm/litre.

for 65Zn and 0. 24 pCi/litre for 54Mn.The adjusted 65Zn concentration shows a plateau in the salinity range 6 to 15%,,, with mean 4. 9 pCi/litre, a difference of 2. 3 pCi/litre from the inflowing concentration.Adjusted concentrations at salinities greater than 20% have mean 12. 6 pCi/litre, a difference of 10. 0 pCi/litre, If these differences were supplied by desorption from incoming suspended particulate 65Zn, knowledge of the concentration of the latter would permit calculation of the percentage desorption.Using the mean adjusted suspended 65Zn concentrations at salinities lessthan 1%0, (12. 7 pCi/litre), to represent this, one finds 19% desorption (2. 3/12. 7 times 100%) in the salinity range 6 to 15% and an average of 79% desorption (10/12. 7 times 100%) at salinities greater than 200%,While only little 65Zn has been found to be desorbed from bottom sediments [ 61, diffusion from the large reservoir of 65Zn in bottom sediments in the estuary and desorption from resuspended bottom sediments might serve as partial sources of the observed difference.This is probably most important in the near bottom (high salinity) water samples.Therefore, 79% is probably an inflated estimate of the desorption of 65Zn from incoming suspended matter. 233

IAEA-SM-158/8 133

TABLE II.RADIONUCLIDE DISTRIBUTION IN COLUMBIA RIVER WATER MIXED WITH FILTERED SEAWATER 30 OCTOBER, 1969 Data are shown as percentages of dissolved plus particulateactivity ± Icy counting errors

Treatment Phase 51Cr 46Sc 65Zn 54Mn

Unmixed Particulate 4.91 0.4 73.41 2.5 57.21 2.5 711 12 Dissolved 95.1 ± 0.9 26.6 f 1.3 42.8 f 2.8 29 t 5

Seawater Particulate 4.1 1 0.3 72.0 t 1.9 31.9 t 2.4 53 7 Mixed Dissolved 95.9 ±0.8 28.0 + 1.1 68.1 t 2.7 47 a 6

Desorption from bottom sediments, either in place or temporarily resuspended, must contribute to the increase in dissolved 54Mn found at higher salinities.The difference between the adjusted mean dissolved 54Mn concentration of samples with salinity greater than 200/60 and the estimated adjusted input concentration is 1.22 pCi/litre (= 1.46- 0. 24).The incoming adjusted suspended particulate concentration is 0. 74 pCi/litre.The percentage desorbed would have been 165%, obviously impossible. An experimental approach which eliminates the ambiguity in the contribution of the sources to the increase in dissolved radionuclide concentrations is desirable.

3. SEAWATER MIXING EXPERIMENTS Estuarine mixing was simulated by mixing seawater with Columbia River water in large containers.Seawater and river water were mixed in equal volumes, resulting in a final mixed salinity of about 16%0.A pratical upper limit for experimental contact times is set by theneed to minimize chemical and biological reactions on the container surfaces.This sets an upper limit of several days, after which periphytic biological uptake or sorption of radionuclides on the container walls would probably be important.This seems sufficient to allow for ion-exchange reactions [ 12] or flocculation reactions [ 13] to take place as they operate on the time scale of minutes or hours. The first mixing experiments were made with Columbia River water collected 31 October, 1969 from behind McNary Dam (see Fig. 1) 125 km downstream from the Hanford reactors.Fifty litres of river waterwere transported to Corvallis in polypropylene carboys.Twenty-five litres were added without filtration to 25 litres of filtered seawater (salinity about 32%o )1.The remaining 25 litres were left unmixed as a control.After 4 h, the mixed and control samples were filtered and the soluble and particulate fractions analysed for gamma-emitting radionuclides.The analytical procedure was as described above except that no stable carriers were added to the filtrates, and the recovery of the dissolved radionuclideswas assumed to be complete.The results are shown in Table II.

I In this and all other samples of seawater used in the mixing experiments negligible concentrations of the radionuclides of interest were found. 234

134 EVANS andCUTSHALL

TABLE III.RADIONUCLIDE DISTRIBUTION IN COLUMBIA RIVER WATER MIXED WITH FILTERED SEAWATER 30 JULY, 1970 All values as pCi/litre ± la counting error.Parentheses indicate percentage of dissolved plus particulate

Treatment Dissolved Particulate Total

Filtered river water 0.30 ± 0.08 lost ----- and seawater 0.43 10.07 0.66 ± 0.5 1.09 ± 0.09 (40"Jo) (601o)

Unfiltered river water 0.48 ± 0.14 0.53 ± 0.04 1.01 t 0.15 and seawater (48%) (52%)

0.85 t 0.09 0.56 ± 0.06 1.41 ± 0.11 (6001o) (400/o)

Control: unmixed 0.14 ± 0.10 0.46 ± 0.07 0.60 ± 0.12 river water (23%) (77%)

0.08 ± 0.05 0.73 ± 0.05 0.81 ± 0.07 (10%) (9010)

Filtered river water 26.1± 1.0 lost ----- and seawater 26.3± 0.9 2.2± 0.1 28.5± 0.9 (93%) (70/0)

Unfiltered river water 26.7± 1.4 2.6± 0.1 29.3t 1.4 and seawater (91%) (9%)

E 31.0± 1.1 2.5± 0.2 33.5± 1.1 (92%) (8%a) A U Control: unmixed 25.1± 1.3 1.4± 0.2 26.5± 1.3 river water (95%) (5%)

26.5 ± 0.7 1.8± 0.2 28.3± 0.8 (94%) (6%)

Filtered river water 0.14 ± 0.06 lost ----- andseawater 0.15 t 0.05 0.75 ± 0.04 0.90 ± 0.06 (17%) (83%a)

Unfiltered river water 0.17 ± 0.07 0.90 t 0.03 1.07 ± 0.08 wI and seawater (15%) (85%) E 0.17 ± 0.04 0.78 ± 0.04 0.95 ± 0.06 (18%) (82%)

Control: unmixed 0.20 ± 0.07 0.81 ± 0.05 1.01 ± 0.09 river water (20%) (80%)

0.25 ± 0.04 0.97 ± 0.04 1.22 ± 0.06 (20%) (80%) 235

IAEA-SM-158/8 135

TABLEIII. (Continued)

Treatment Dissolved Particulate Total

Filtered river water 0.35 t 0.05 lost ----- and seawater 0.36 ± 0.04 0.05 f 0.02 0.41 ± 0.05 (88%) (12%a)

Unfiltered river water 0.21 : 0.05 0.01 ± 0.01 0.22 ± 0.05 and seawater (95%) (5%)

0.17 t 0.03 0.03 ± 0.02 0.20 ± 0.05 (85%) (15%)

Control: unmixed 0.31 ± 0.06 0.00 ± 0.02 0.31 ± 0.07 river water (100%) (0%)

0.34 t 0.03 0.00 ± 0.02 0.34 ± 0.04 (100%) (0%)

There was no significant difference in the distribution of 51Cr and 46Sc between the seawater mixed and control samples. 5lCr remains mostly dissolved (about 95%) while 46Sc remains mostly particulate (about 75%). In contrast to this conservative behaviour,65Zn is relatively more soluble in the seawater mixed sample (68%) than in the control sample (43%).This suggests that contact with seawater has caused the release of a significant fraction of the particulate associated 65Zn.The fraction of the 65Zn desorbed can be calculated: fraction "de sorbed" _ (57% - 32%)/57% = 0.44. Counting errors make the apparent 25% desorption of 54Mn insignificant at the 95% confidence level, as determined by a t-test. A second mixing experiment of this type was performed in the field on 30 July, 1970.Surface water was pumped from the Columbia River in the main channel, opposite Harrington Point (see Fig. 2) and upstream of the limit of intrusion of salt water into the estuary.Duplicate samples of 50 litres each were filtered and mixed with an equal volume of filtered seawater in 55-gal (US) barrels lined with polyethylene bags.Duplicate samples of unfiltered water were similarly mixed with filtered seawater.Duplicate control samples of unfiltered river water were collected but not mixed with seawater.All samples were allowed to stand for 24 h reaction time and then filtered.Analysis for the radionuclides present in each phase was as before; stable carriers of zinc, chromium, and manganese were added this time.The radionuclides 65Zn, 51Cr, 46Sc, and 124Sb were in quantitatively measurable concentrations. The mixing of filtered river water with filtered seawater was designed to test the possibility of flocculation or precipitation from solution of radionuclides in the Columbia River estuary.Because both river-water and seawater samples had been filtered before mixing, the detection of any radionuclides on filters after subsequent filtration could be attributed to the formation of new particulate phases. 236

136 EVANS and COTSHALL

No significant concentrations of any radionuclide were found on the filters after this second filtration, indicating the absence of flocculation or precipitation.It was felt, however, that the centrifugal pump used to force the samples through the filters might have broken apart flocculant aggregates, allowing them to pass through the filters and thus go undetected.Therefore, this experiment was repeated with samples collected 26 August, 1971 using air pressure (3 atm) to force the water sample through the filters.Again, no significant activity was found on the filters, confirming the earlier result. The mixing of unfiltered river water samples with filtered seawater had the same goal as the earlier mixing experiment of 31 October, 1969.The results are shown in Table III. There was little change in the distribution of 51Cr, 46 Sc, and 124Sb between the soluble and particulate phases when compared to the control. These radionuclides would appear to be conservative. In contrast, 65Zn was markedly more soluble in the mixed system (54%) than in the control (17%).It would appear that about 450/, of the 65Zn associated with the particulate phase was transferred to solution.This conclusion is clouded by the lack of a constant total of65Znactivity among the systems.The .lower total 65Zn activity of the control could be due to sorption of soluble 65Zn on the polyethylene container surface, thus biasing the result to indicate an apparent "desorption" from particles. 3. 1.Seawater leaching of suspended matter Seawater leaching of Columbia River suspended matter separated from river water allows a more sensitive determination of the percentage desorption of the radionuclides.Suspended matter collected on filters as before was placed in 1-litre polyethylene containers to which 0. 5-litre of filtered seawater was added.The seawater was filtered from individual sample containers after progressively longer contact times.The solutions were evaporated to dryness, and the sea salts packed in counting tubes for analysis.The leached suspended matter and accompanying filterswere digested and analysed as before.Table IV shows the percentage of the

TABLE IV. PERCENTAGE OF COLUMBIA RIVER RADIONUCLIDES LEACHED FROM SUSPENDED MATTER BY SEAWATER CONTACT

Collection date Seawater contact time Total activity in seawater fraction (°1o)

65Zn 54Mn 51Cr 46SC 1h 3f 4 5t 9 0t 3 2t 2 30 July 4 h 5 f 4 18 t 7 0 t 2 7 t 2 1970 12h 16f 3 18 5 01 6 0t 4 1 week 41 ± 3 64 1 5 0 t 5 2 t 2 1 week 30 ± 3 55 ± 5 0 t 7 3 f 2 1 week 36 ± 4 50 ± 6 0 t6 3 t 2

6 January 3 weeks 1970 14 ± 3 29 ± 13 1 t 2 1 t 2 237

IAEA-SM-158/8 137 total activity of individual radionuclides found in the seawater fraction for samples collected 30 July, 1970 and a single sample collected 6 January, 1970.Included are uncertainty estimates calculated from the one standard error counting errors. Again 65Zn and54Mn contrast with 51Cr and 46Sc.No significant desorption of the latter two takes place. A higher percentage of 54Mn than65Zn was desorbed.It is of interest that very little 65Zn was desorbed within 4 h contact time, suggesting that ion exchange, which is rapid,is not the cause of 65Zn desorption. An odor of decay in the one-and three-week samples suggests the possible role of microbial degradation.

3. 2.Seawater leaching of bottom sediments The observed desorption of 65Zn from suspended sediments stands in contrast to the results of Johnson et al.[ 6], who found little desorption of65Zn from bottom sediments in the Columbia River.One possible explanation for these differences might be that the 1-h seawater contact timeused was too short. To determine if longer contact times might result in increased desorption of65Zn from bottom sediments, samples of bottom sediments (silty- sand) were collected from behind McNary Dam and transferred to Yaquina Bay.About 5 kg of sediment were placed in a plastic basin to a depth of about 8 cm, covered with a plastic screen to exclude larger organisms, and suspended in the water column from the OSU Marine Science Center dock.Sediment was also sealed in 25-cm lengths of dialysis

TABLE V. RADIONUCLIDE CONCENTRATIONS IN COLUMBIARIVER BOTTOM SEDIMENTS TRANSFERRED TO THE SEAWATERENVIRON- MhNT OF YAQUINA BAY Concentrations are in unitsof pCi/g ±1a counting errors

Seawater contact time 60Co 65Zn 54Mn 46Sc (d)

3.7 0 7.0 ±1.3 69.7 ±11.9 4.4 ±2.1 29.4 ± 4.3 0 6.6 ±1.2 71.5 ±11.1 4.0 ±1.9 16.0 ± 0 6.8 ±1.3 84.4±12.9 4.3 ±2.2 33.8 ±3.7 3.0 0 7.0 ±1.2 70.1 ±11.4 4.5 ±2.0 22.2 ± 3.2 5 7.0 ±1.5 69.9 ±11.6 4.4 ±2.0 20.4 ± 5a 6.5±1.2 75.1±11.4 4.3±2.0 21.2±2.8 12 6.8 ±1.1 74.1 ±10.6 3.4 ±1.7 21.0 ±3.1 12a 6.5 ±1.1 75.3 ±11.1 2.5 ±1.7 20.7 ±3.2 26 6.3±1.2 68.6 ±12.0 2.4 ±1.9 27.0 ±6.0 26a 6.9 ±1.2 75.1 ±11.7 2.8 ±1.9 27.0 ±6.0 26a 6.7 ±1.3 70.4 ±12.1 2.6 ±1.9 21.9 ±6.5 50 6.3 ±1.1 66.3 ±10.4 3.2 ±1.7 19.5 ±4.0 50a 6.7 ±1.2 69.2 A.11.1 2.2 ±1.7 16.6 ±3.8 50a 6.2 ±1.1 70.6 ±10.3 2.1 ±1.6 19.5 ±3.9 77a 6.4 ±1.2 74.7 ±11.5 1.9 ±1.7 17.9 ±4.9

a Samples in dialysis tubing 2 38

138 EVANS and CUTSHALL

tubing and submerged along with the sediments in the plastic basin. Samples were recovered at intervals over an eleven-week period, dried, weighed, packed into counting tubes, and radioanalysed.Salinity of the bay water was determined from surface samples collected on days 12 and 26; it was 33. 7 and 31. 5%,, respectively. Table V shows the results of these analyses, corrected for decay to the date of submersion. 85Zn and all other radionuclides except 54Mn remained essentially unchanged in concentration.About half the 54Mn was lost by the eleventh week, confirming the results of Johnson et al.[ 7 ] . It appears that the reaction with seawater differs for suspended sediment and bottom sediment, at least for 65Zn.

4. SUMMARY AND CONCLUSIONS (1) The radionuclides 65Zn and 54Mn in the Columbia River are partially desorbed from suspended particulate matter during estuarine mixing.The calculated estimates of percentage suspended 54Mn and 65Zn desorbed varied over a wide range.This is probably the result of analytical uncertainties, the experimental procedures used, and temporal variations in the chemical and biological character of seawater and Columbia River water.The percentage desorption of 65Zn ranges from 15 to 45% while that of 54Mn is higher, 30 to 60%. (2) There was no evidence of desorption of 51Cr, 46Sc, and 124Sb. (3)None of the above radionuclides showed any evidence of removal from solution as the result of flocculation, precipitation, or absorption consequent on mixing with seawater within the estuary. (4)Desorption of 65Zn from suspended particles by seawater is greater than desorption of 65Zn from river-bed sediments. The differing behaviour of the radionuclides studied may be related to their ionic states in solution in the Columbia River.Nelson et al.[ 141 reported that 65Zn and 54Mn exist largely as cations (83 and 48%, respectively) in samples nearest the estuary.Anionic forms predominate for 51Cr (99%) and 46Sc (79%) while 124Sb was reported to exist mostly as an un- charged species (84%).The cation exchange capacity of soils and clay minerals is generally much greater than their anion exchange capacity [ 12]. Cationic 65Zn and 54Mn would be expected to participate more readily in exchange reactions with the ions in seawater. It may be interesting to reflect upon the above treatment of salinity data.Salinity is used as an indicator of the degree of dilution of river water with seawater in estuarine samples. On the other hand salinity may also be considered as an indication of the abundance of potentially reactive ions derived from seawater.This duality is probably acceptable so long as interactions other than those considered do not affect the conservation (i. e,linearity of mixing) of salinity and reactant species.While our approach is founded upon the assumption that the most likely reactant is a major constituent of seawater (e. g. Mg),salinity changes during mixing are accompanied by changes in other parameters such as temperature, biological activity and particle electrochemistry [15].Variations in the action of these parameters may well be non-linear with salinity. We have not characterized the specific mechanism causing changes in the soluble/ suspended partitioning of 65Zn and54Mn. 239

IAEA-SM-158/8 139 ACKNOWLEDGEMENTS Appreciation is extended to Peter J. Ilanson in generously allowing us to use data he collected, to Norman Kujala, skipper of R/V"Sacajawea", and to Vernon G. Johnson for frequent discussions and commentary. The senior author acknowledges financial support from the National Science Foundation and the State of Oregon.

REFERENCES

[1] FUKAI, R., in Disposal of Radioactive Wastes into Seas, Oceans, and Surface Waters (Proc.Symp. Vienna, 1966), IAEA, Vienna (1966) 483. [2] KHARKAR, D.P., TUREKIAN, K.K., BERTINE, K.K., Stream supply of dissolved silver, molybdenum, antimony, selenium, chromium, cobalt, rubidium and cesium to the oceans, Geochim.cosmochim. Acta 32 (1968) 285. [3] DE GROOT, A.J., DE GOEIJ, J.J. M., ZEGERS, C., Contents and behaviour of mercury ascompared with other heavy metalsin sedimentsfrom the riversRhine andEms, Geologic Mijnb. 50 3 (1971)393. [4]LOWMAN, F.G., PHELPS, D.K., McCLIN, R., ROMAN DEWEGA, V., OLIVER DE PADOVANI,I., GARCIA, R.J., in Disposal of Radioactive Wastes into Seas, Oceans and Surface Waters (Proc.Symp. Vienna, 1966), IAEA, Vienna (1966) 249. [5] PERKINS, R. W., NELSON, I.L., HAUSHILD, W.L.. Behaviour and transport ofradionuclides in the Columbia River between Hanford and Vancouver, Washington, Limnol. Oceanogr. 11 (1966) 235. [6] JOHNSON, V., CUTSHALL, N., OSTERBERG, C., Retention of 65Zn by Columbia River sediment, Water Resources Res. 3 (1967) 99. [7] CUTSHALL, N., JOHNSON, V., OSTERBERG, C., Chromium-51 in seawater: chemistry, Science152 (1966) 202. [8] CONOMOS, T.J., GROSS, M.G., Mixing of Columbia River and ocean waters in summer, J. sanit. engng. Div. Am. Soc. civ.Engrs 945 (1968) 979. [9] PATTULLO, J., DENNER, W.,Processesaffectingseawater characteristicsalong the Oregon coast, Limnol. Oceanogr. 10 (1965) 443. [10] DRAPER, N., SMITH, H., AppliedRegressionAnalysis, Wiley, New York (1966). [11] HANSON, D.J., Vertical distribution of radioactivity in the Columbia River estuary, Master's thesis, Corvallis, Oregon State University (1967), 62 numb, leaves. [12] WIKLANDER, L., "Cation andanion exchangephenomena". Chemistry of the Soil (BEAR, F.E., Ed.), Am. Chem. Soc. Monograph No. 160, Reinhold, New York (1964) Ch.4. [ 131 HARRIS, R. W., "Hydraulics of the Savannah Estuary", Studies of the Fate of Certain Radionuclides in Estuarineand other Aquatic Environments (SABO, J.J., BEDROSIAN, P.H., Eds), U.S. Public Health Service Publication No. 999-R-3, Washington (1963). [14] NELSON, J.L., PERKINS, R.W., NIELSEN, J.M., in Disposal of Radioactive Wastes into Seas, Oceans, and Surface Waters (Proc. Symp. Vienna, 1966), IAEA, Vienna (1966) 139. [15] PRAVDIC, V., Surface charge characterization of sea sediments, Limnol. Oceanogr. 15 (1970)230.

DISCUSSION

M.BERNHARD:Could you describe some characteristics of the Columbia River sediments? D. W. EVANS:The Columbia River drains the Columbia Plateau, which is largely composed of basaltic rock.The bottom and suspended sediments are notout oftheordinary,being composed predominantly of quartz, feldspars,and various clayminerals,such as montmorillonite. The suspended sediments, being of a finer granularcomposition,include a higher percentage of clay minerals than the bottom sediments. 240

140 EVANS and CUTSHALL A point to be noted is that, because the Columbia River drains a region of basic rocks, its pH is higher than that of other typical rivers.I do not think that the small change in pH from 7. 5- 8. 0 in the Columbia River to about 8. 0 in north-east Pacific Ocean water is the cause of desorption of 65Zn and 54Mn. The slightly increasing pH would favour increased adsorption rather than desorption of these radionuclides. J. L. GLENN: Were the samples from the estuary collected at about the same time?If not, how were the observed radionuclide concentrations adjusted to obtain an instantaneous picture? D.W. EVANS: The samples from the first group of stations were collected during a 3-h period.The samples along the vertical profile were collected during a 25-h period. No adjustment was made to obtain an instantaneous picture.It is possible that the concentrations of radionuclides entering from the river changed during the sampling period.Ideally, one would like to sample from a single parcel of Columbia River water as it mixes with ocean water in the estuary, but this was not done, and I do not know how it could be done. B. PATEL: What is the organic content of Columbia River sediment, and did you find any correlation between the amount of organic matter and the uptake of various nuclides?In studies on Bombay harbour sediment we found that the organic content did influence the uptake of nuclides. D. W. EVANS: The organic content of bottom sediments and suspended particulate matter was not determined concurrently with the experiments described here. The organic content of Columbia River bottom sediments varies considerably.It may be as much as 2 to 5% in fine, clay-sized fractions, but very much less in coarser, sand-sized fractions. The relationship you noted between organic content and radionuclide uptake may actually reflect the relatively greater adsorption of radionuclides on the finer sediment fractions.This has been observed by a number of workers, and organic content is only incidentally associated with this greater radionuclide uptake. H. PARKER: Can you tell us why you filtered the seawater in the mixing experiments, and estimate how the results might have been affected had you not done so? D. W. EVANS: The mixing experiments were designed to simulate mixing in the estuary.At the same time we wanted to isolate desorption effects from any possible simultaneous adsorption effects on the suspended particulate matter brought into the estuary by ocean water.In any event, the concentrations of suspended particulate matter in ocean water were very much less than in the Columbia River water, so that adsorption on this particulate matter was probably negligible. H. PARKER: Would it not have been less ambiguous to imitate as closely as possible what actually happens in the estuary'? D.W. EVANS: Yes, but the effects of desorption or adsorption could not then have been determined in isolation. 241

RLO-2227-T12-29

Marine Chemistry Elsevier Publishing Company, Amsterdam-Printed in The Netherlands

CALCIUM-MAGNESIUM RATIOS IN THE TEST PLATES OF ALLOCENTROTUS FRAGILIS

JAMES L. SUMICH and J. E. McCAULEY Grossmont College, lil Cajon, Calif. (U.S.A.) Oregon State University, Corvallis, Oreg. (U.S.A.) (Accepted for publication June 27, 1972)

ABSTRACT

Sumich, J. L. and McCauley, J. E., 1972.Calcium-magnesiumratios in the test plates ofAllocentrotus fragilis. Mar. Chem., 1: 55-59.

Calcium-magnesium ratios in test plates of the sea urchinAllocentrotus fragilis(Jackson) are age-dependent and not temperature-dependent.

INTRODUCTION

The skeletal material of echinoderms is comprised primarily of CaCO3 with lesser amounts (usually less than 15 %) of MgCO3. The variation of the Mg concentration in echinoderm skeletons has been studied for over a half-century, yet the underlying causes of these variations are still not understood. In a recent review, Dodd (1967) listed six factors which he believed to be important in determining the skeletal Mg composition of marine organisms: (1) the amount of Mg relative to Ca in sea water and in body fluids; (2) spatial isolation of the calcification site from sea water; (3) taxonomic level of the organism; (4) ontogeny of the organism; (5) growth rate of the organism; and (6) other environmental factors such as water temperature and salinity. The difficulty in obtaining complete and precise information concerning some or all of these conditions at the time and site of collection has been the major drawback of previous studies of skeletal Mg. To minimize or eliminate some of these variables, the skeletal Mg content of a single species of echinoid,Allocentrotus fragilis(Jackson), was studied. This species has a narrow temperature range and its age can be determined accurately. It occurs alongthe Pacific Coast of North America in depths of 50 to 1,260 in, spanning a temperature rangeof 10.5°C to 3.5°C. Seasonal temperature fluctuationsaresmall, less than 2°C at 200 m, and almost non-existent below 400 in. These temperatures have been inferred from hydro- graphic data of the Oregon State University Department of Oceanography and verified by direct measurements by Dr. A. G. Carey Jr. (unpublished data). 242

56 J. L. SUMICH AND J. E. McCAULEY

We recently reported a simple age determination method forA. fragilis,using visible growth zones in test plates (Sumich and McCauley, in preparation). Growth rates at 800 and 1,260 m were noticeably less than at 200 m. Clarke (1911) and Clarke and Wheeler (1917) were the first to report a positive correlation between the Mg concentration of echinoderm calcite and environmental water temperature. In 1922, Clarke and Wheeler demonstrated a relationship between skeletal composition (especially Mg concentration) and the taxonomy of certain invertebrate groups. Eighty-two echinoderms were analyzed, of which only 14 were echinoids. Chave (1954) reviewed the literature on magnesium in calcareous marine organisms. Pilkey and Hower (1960) studied the intraspecific variation of skeletal Mg inDendraster excentricus(Eschscholtz), a Pacific Coast sand dollar. Again, a direct correlation between water temperature and skeletal Mg concentration was found. At least forD. excentricus, the variation of skeletal Mg concentration was much less at the intraspecific level than at the interspecific or class level. Variations of echinoderm skeletal Mg concentration have been reported at many levels of systematic organization. They have been shown to exist within a single skeletal unit (tooth), between different skeletal parts of the same specimen, and between homologous skeletal components of different echinoid species from the same marine community (Raup, 1966; Weber, 1969). Raup (1966) confirmed the correlation between skeletal Mg concentration and environmental water temperature, but stressed that the large variations observed could not be explained by water temperature fluctuations alone. The skeletal Mg variation between different parts of an individual is often as great as the class-level variations attributed by other workers to temperature effects. Allocentrotus fragiliswas collected from depths of 200, 800 and 1,260 m. The tests were cleaned, then dissolved in 5 % HCI. The Mg concentrations were determined using atomic absorption spectrophotometry (Perkin-Elmer Corp., 1968). Perivisceral fluids of specimens from 200 and 800 m were frozen at the time of collection and later analyzed for Ca and Mg. Perivisceral fluids of specimens from 1,260 m were not available. The skeletal MgCO3 content of one trawl collection ofA. fragilis fromeach depth is shown in Table I.

TABLE I

Summary ofskeletal Mg content of A. fragilis

Depth Temp. No. of Percent skeletal MgCO3 (m) Co specimens mean + S.D. range

200 8.0 23 7.2+0.4 6.4-7.9 800 4.3 18 7.0+0.3 6.6-7.6 1260 3.5 5 7.3+0.3 7.1-7.7 CALCIUM-MAGNESIUM RATIOS INALLOC/iNTROTUS ERAG/L/S 57

No correlation between the skeletal Mg content and environmental water temperature is apparent from the data presented in Table 1. However, when the skeletal Mg values of A. fragilisat each depth are plotted against the urchin's age, a definite trend develops (Fig.1). The regression lines of skeletal Mg content for each depth indicate that either the amount of Mg incorporated into newly-formed skeletal calcite decreasesas the urchins grow older; or that Ca ions gradually replace Mg ions after the calcite is formed. If the former were true, older skeletal test platesnear the peristome (Sumich and McCauley, in preparation) should contain more Mg than the newer plates whichare near the apical area. However, if Ca was gradually substituted for Mg in the calcite lattice, older plates should contain equal or lesser amounts of Mg than younger plates.

200 m S.E. 0.20 8.0 - - - 800m S.E. 0.20 1260m S.E. 0.14

1 70I-.. o\°

60i_

___J------J 0 2 1--13 4 5 6 Age (years)

Pig.l. Variation ofA. fragilis skeletal MgCO, with age at 200, 800, and 1,260 m depths.

That older A. fragilis incorporate less Mg into the new skeletal calcitewas shown by a series of Mg determinations along the length of complete interambulacral test plate columns (Table II). Test plates of a 66 mm specimen (6 years old) from 200m and of a 27 mm specimen (2 years old) from 1,260 m were used.

TABLE II

Percent MgCO, of A. fragilis test plates along an interambulacralcolumn.Plates are numbered from the peristomialmargin,and treated in groups of three to facilitate analysis

Depth (m) Plate no. 1 -3 (oldest) 4-67-9 10-12 13-15 16-18 19-21 (newest)

200 7.1 6.9 6.5 6.4 6.4 6.2 5.9 1260 8.3 7.1 6.4 5.9 6.0

A definite decrease in Mg content from the oldest to the newest plates in conjunction with data shown in Fig.1 indicates that, asA. fragilisages and the growth rate decreases, less Mg is incorporated into the skeletal material being formed. Dodd (1967) and Weber (1969) have shown that echinoids with slower growth rates generally have smaller amounts of skeletal Mg. Weber (1969) proposed that Mg is incorporated into the calcite lattice by random substitution for Ca at the site of calcification. The Mg ion, however, is less stable than the Ca ion in the crystal lattice. Thus, in slowly forming calcite (viz., 244

58 J. L.SUMICIIAND J. E. McCAULEY

slowly-growing urchins) more Mg may be excluded from the calcification site by the more stable Ca ion prior to actual calcite formation. Conversely, in rapidly-forming calcite more Mg may be incorporated into the calcite lattice. Preliminary Mg and Ca analyses of the perivisceral fluids ofA. fragilissuggest an additional mechanism for the formation of high Mg calcites in younger urchins. The Mg and Ca concentrations in the perivisceral fluids of 18A. fragilisfrom 200 m exhibit a negative correlation with the urchin's age (Fig.2). Analyses of the perivisceral fluids of

Mg at 800m

Mg S.E. 161 1200 0_j0 0 - -Co S.E. 62 0 0 0 800 0 0 9 0 Cc at 800M

oci 400

0 2 3 4 5 6 7 Age (years)

Fig.2. Mg and Ca concentrations in perivisceral fluids of A. fragilisfrom 200 m.Mean values and ranges of Ca and Mg at800 m are also shown.

15 A. fragilis from 800 m were also made. However, all these urchins were young and near the same age (2 years old). The range of perivisceral fluid Mg and Ca concentrations from 800 m are also shown in Fig.2. These values are only slightly greater than the Mg and Ca values of A. fragilis of the same age from 200 m. The Mg and Ca concentrations in the body fluids of the youngest urchins were very similar to values reported by Sverdrup et al. (1942) for sea water (1,278 p.p.m. and 400 p.p.m., respectively). The greater Mg concentration in the perivisceral fluids of young urchins may, in part, account for the high Mg concentrations found in the test plates of these urchins. Although echinoids are known to vary the Ca content of their perivisceral fluids by over 50% with respect to sea water (Binyon, 1966), no mechanisms have been proposed to explain age variations of Mg and Ca in the perivisceral fluids of A. fragilis. The deposition of magnesium in the calcite of the regular sea urchin, Allocentrotus fragilis, is age-dependent rather than temperature-dependent. This research was partially supported by a contract with the Atomic Energy Commission AT(45-1)2227 Task Agreement 12. It is assigned manuscript number RL02227-T 12-29.

REFERENCES

Binyon, J., 1966. Salinity tolerance and ionic regulation. In: R.A.Boolootian(Editor),Physiology of Echinodermata.Interscience, New York, N.Y., pp. 359-377. Chave, K. E., 1954. Aspects ofbiochemistry ofmagnesium.1. Calcareous marineorganisms.J. Geol., 62: 266-283. 245

CALCIUM MAGNESIUM RATIOS IN ALLOCIENTROTUS FRAGILIS 59

Clark, A. H., 1911. On the inorganicconstituents of the skeletons of two Mus. Proc., 39: 487. recent crinoids.U.S. Natl. Clarke, F. W. and Wheeler, W. D.,1917. The inorganic constituents invertebrates. of marine constituents of marine U.S. Geol. Surv., Prof Pap.,102: 56 pp. Clarke, F. W. and Wheeler, W. D., 1922. The inorganic constituentsof marine invertebrates. Surv., Prof. Pap., 124: 1-62. U.S. Geol. Dodd, J. R., 1967. Magnesiumand strontium in calcareous skeletons: 41: 13/3-/329. a review..1.Paleontol., Perkin-Elmer Corporation, 1968. Revisions ofA nalytical Methods forA tomic A bsorption Spectro- photometry.Norwalk, Connecticut, not paginated. Pilkey, O. H. and Hower, J., 1960. The effect of environmenton concentration of skeletal magnesium and strontiuminDendraster. J. Geol.,68: 203-216. Raup, D. M., 1966. The endoskeleton. In: R. A. Boolootian (Editor),Physiology of Echinodermata. Interscience, New York, N.Y.,pp. 379-395. Sumich, J. L. and McCauley, J. E., in preparation. Growth ofa sea urchin, the Oregon coast. Allocentrotus fragilis,off Sverdrup, H. V., Johnson, M. W. and Fleming, R. H., 1942.The Oceans. Their Physics, Chemistry,and General Biology.Prentice-Hall, Englewood Cliffs,N.J., p. 166. Weber, J. H., 1969. The incorporationof magnesium into the skeletal calcite Sci., 267: 537-566. of echinoderms.Am. J. 246

RLO-2227-T12-35

A systematic review of the rattail fishes(Macrouridae:Gadiformes) from Oregon and adjacentwaters. By Tomio Iwamoto and David L. Stein.0cc. Pap.azlif.A,ad. Sci. (in press.)

PRECIS

Nine species of macrourids are reportedto inhabit waters off the Pacific coast of North America between San Francisco, California and Alaska. They are: Coryphaenoides acrolepis (Bean), C. armatus (Hector), C. cinereus (Gilbert), C. filifer (Gilbert), C. leptolepis(GUnther), C. Longifilis (GUnther), C. pectoralis (Gilbert),C. yaquinae (new species), and Nezumia stelgidolepis (Gilbert).All nine species are described and illustrated and a key (which includes two additional,more southerly species, Coelorin- chus scaphopsis (Gilbert) and Nezumialiolepis (Gilbert) to adult specimens is provided.Generic problems associated with Coryphaenoides, Chalinura, Nematonurus and Lionurus are discussed, and characters suchas dentition, body proportions, number of precaudalvertebrae, squamation on the head, and number of retia are considered in their roleas generic characters. We tentatively follow Okamura's (1970;1971)treatment of Coryphaenoides and treat Nematonurus and Chalinuraas subgenera of Coryphaenoides. The divergent position of Coryphaenoidespectoralis is noted and the possible position of the species ina separate sub-genus or genus is suggested. 21,7

RLO-2227-T12-36 Accepted byHealth Physics

TURNOVER OF 65Zn IN OYSTERS

Norman Cutshall School of Oceanography Oregon State University Corvallis, Oregon 97331

Abstract--The uptake and lossof 65Zn by Pacific oysters are fitted by solutions to a first-order, lineardifferential equation that describes "single-compartment" isotope substitutionkinetics.The rate constants for loss are related to the rateconstants for uptake and to steady- state concentration factors for stablezinc and 65Zn. The single-compartment solutionsadequately fit uptake and loss data from field experimentsextended as long as several hundred days. The rate constant estimated fromloss data is 5.49 X 10-3 d-1 and that from uptake data is 4.93 X 10-3 d-1.The apparent importance of multiple compartments can be affectedin experiments where uptake times short. are

Wolfe(') remarkedupon the notoriety of oysters as accumulators of zinc and showed that the metalis predominantly bound to soluble, high molecular weight proteins.Because of their role as concentrators of zinc and their commercial importance,oysters have drawn attention in studies of radioactive 65Zn (T1/2= 245 days).Rapid accumulation of 65Zn byoysters immersed in seawater containingthe isotope was noted in early radioecological literature (2,3).In the 1950's 65Zn was reported in Pacific oysters (Crassostreagigas)and other organisms collected along the coast of Oregonand Washington(45). Zinc-65 was introduced into this region by theColumbia River, whose waters had been used to cool reactors at Hanford,Washington.

Several studies have been devotedto the kinetics of accumulation and loss of 65Zn by oysters. Chipman(3) noted rapidaccumulation rates. Seymour(6) reportedon accumulation and loss of this nuclide by oysters in a natural environment. Salo and Leet(7)related rates of accumulation of 65Zn by oysters ina reactor effluent canal to reactor discharge rates. Rcmeril(8) determinedrates of uptake of 65Zn for variousor- gans of oysters in the laboratory, and Wolfe(1)reported changes in F'5Zn contentof oysters during times when the 65Znsupply (worldwide fallout was thesource in his case) was diminishing. Each of these authors used some form of exponentialequation to fit his data. The adequacy of exponential modelsto describe kinetics in complex ecological situations has been questioned(9r10). The latter authors suggest 248 that power functions more adequately describe radionuclide kinetics in marine organisms. The exponential and power function models were contrasted by Feldt(ll) in keynoting a recent marine radioecology conference. Reichle et al.(12) noted that solutions based on power functions provided better estimates of biological turnover for bone-seeking radionuclides than did exponentials. Many experiments designed to measure biological turnover rates of radionuclides emphasize only loss data; whereas, uptake data are, in principle, equally useful. Indeed, comparison of rate parameters determined both by uptake and loss experiments provides a test of the adequacy of a single-compartment kinetic model and its exponential solution. Consequently, we have re-examined data reported by Seymour'6) in order to determine the adequacy of single- compartment kinetics in describing 65Zn turnover in oysters and to show the relationship between uptake, loss, and steady-state.

Briefly, Seymour's experiment consisted of cross-transplants of oysters(Crassostrea gi as)from Willapa Bay, near the Columbia River mouth, to essentially 6 Zn-free waters in Puget Sound of Hood Canal, and vice versa. The loss of 65Zn from oysters taken from Willapa Bay was followed in time, as was the accumulation of radionuclide by oysters planted in Willapa Bay. Both loss and accumulation were followed for as long as two years following transplant. Although the season of transplant varied and the initial 65Zn concentrations in oysters taken from Willapa Bay varied, the pooled data from eight such groups followed an exponential decline rather well (Seymour(6), Fig, 5), The effective half-life found was 135 days, and the corresponding biological half- life was 300 days. No analysis of the accumulation data was presented, although it was noted that uptake rates were most rapid during early times and that initial uptake rates varied with season of transplant.

Single-Compartment Equations. The equations used below have been often reported. They are reviewed here simply to indicate my procedures and to demonstrate the common source of equations for accumulation, loss, and steady-state conditions. The physical model is one of isotope substitution. With the assumption that an organism's uptake and loss of the total element are equal, its radionuclide dynamics are described by:

dA= RSs- RSb - XA, (1) dt where R =therate of uptake (equal to loss) of all isotopes ofthe element in question per unit body weight; Ss =thespecific activity of the radionuclide in the source; Sb =thespecific activity of the radionuclide in the organism; X =theradionuclide decay constant; and A =theconcentration of radionuclide in the organism. 21.

We have implicitly assumed that the radioisotope makes no significant contribution to the total concentration of the element in the system. This is almost always the case for 65Zn. Still further simplifying assumptions are made before obtaining solutions to equation (1): 1. The source specific activity and elemental uptake rates are constant in time. 2. All atoms of the element in question, contained in the organism, are equally likely to be lost by excretion. 3. No allowance is made for growth of the organism. The mean residence time, or turnover time, of the element in the organism can be defined: T = P- , R where P = total element concentration.The turnover time is related to the more commonly used "biological half-life" (Tl/2b) and to the "biological rate constant" (a):

T =Tl/2b 1

Thus, after substitution, Eq.(1) can be written as

dA= RS -(a + 8)A (2) dt s Solutions may now be readily obtained: Loss Situation (Ss = 0, boundary condition A = AO at t = 0i

A = A . (X + 0) t (3) 0 e- ; Uptake Situation (boundary condition A = 0 at t = 0),

RS + t) (4) The equation for the steady-state condition, dA/dt = 0, may be obtained either by setting equation (2) equal to zero, or by allowing time to become infinite in equation (4): RSs A (steady-state)_ (5) + At steady-state the specific activity of radionuclide in the organism (Sb) is less than the specific activity of the source:

= . o (6) Sb X+s Ss 250

This observation has been qualitatively made before (13),as has the equivalent statement for concentration factors(14). If the radionuclide source is water, then equation (6) can be rearranged to yield:

CF* _ S CF (7)

where CF* is the radioisotope concentration factor, and CF is the elemental concentration factor. Equations (6) and (7) afford a third means for calculating turnover rate constants, if the appropriate specific activities or concentration factorsare known.

Evaluation of Rate Constants. Seymours'(6) loss data were fitted with an equation of the form of (3) using the method of leastsquares (Fig. la). The value of (A + S) thus obtained was 5.71 X 10-3d-1o Differences between the fitted line and the data (residuals)are plotted versus time in Fig. lb. Convergence of the residuals implies that a better fit would result if the datawere weighted(15). Consequently, I fitted the data using the weighting factor 1/A (Fig. la). The value of (A + s) was changed to 5.49 X 10-3 d-1. No pattern is apparent in the plot of the weighted residuals (Fig. lc). Therefore, the latter value for (A + s) is taken as the best estimate from the loss data.

Treatment of the accumulation data is slightlymore complicated in that two parameters must be determined: the effective rate constant (A + $) and the asymptotic value of 65Zn concentration,RSs/(A + S)(Eq. 4). Data from oysters planted in Willapa Bay, fitted by leastsquares, yielded values of the effective rate constant and the asymptote of 4.93X 10-3 d-1 and 215 pCi/g, respectively. The curve for these parameter values and the accumulation data are plotted in Fig. 2a0 Deviations from the fitted points (residuals) are plotted versus time in Fig. 2b. No weighting of data was required for a better fit of the linein Fig. 2a.

Seymour(6) reported concentrationfactors for 65Zn and Zn in Willapa Bay oysters as 9900 ± 2700 and 14000 ± 9700, respectively. Solving equation (7) for (A + $) using these data yields 9.73 ± 7.24X 10-3 d-1.

Discussion. The rate constants determined from uptake data do not differ appreciably from those determined fromloss data. These values do not differ appreciably from the values estimated usingconcentration factors, although the large error term associatedwith the latter makes the comparison relatively insensitive. Residuals from the fitting of uptake and loss data with exponential equations donot reveal any failure of the single-compartment model; that is,no systematic pattern is evident when the residuals are plotted againsttime. 251

8 I

U 8

4000I \ 00 0C a 00 C 0 0 (1) U 0C V 200 r N 0 0 I% N 0 to 0 W 0 z 0 0 o° °00 0 '0044 0 I -8 200 400 600 Time (days)

Figure 1 (a).Analysis of Loss Data:65Zn concentrationvs. time.Solid line is fittedby least-squares using unweighted data; dashedline is fitted using weighting. +300D

8 +200°r 0 0 ° 0 0 + 100I0 0000 0 1 o 0 0 o V 0 0°° 0O 0 Op °-° °° °- ° oo 00 00 0 000 o I 0cc01 0 0 - 100 O°o °°GO 00 0 -200-

L I I 1 0 200 400 600 Time (days)

Figure 1 (b).Analysis of Loss Data:Residuals from unweighted fit vs. time, showing convergence. +2

0 0 0

0 U 0 0 0 0 0

I 1 I 200 400 600 Time (days)

Figure 1 (c)Analysis of Loss Data:Weighted residuals from weighted fit. 300

0 E 0 0 U 0 0 0 Q- 200 0 0 0 0 C 0 00 0 0 4- 0 8o 0 00 0 4-c 0 0 0 0 0 0 C 0 O 0 0 0 100 cP O U 0 c N 0 to c0 09 coo 0 0(5 I I I -i- 0 200 400 600 Time (days)

Figure 2 (a) Analysis of Uptake Data: 65Zn concentrationvs. time. +100

0 0 0 0 +50 000 0 00 0 0 0 00 0 00 0 000 0 o o0 0 0 0 0 0 0 bSo o 8 0 0 0 8 0 - 50 0

I 1 -1001 ( I I 0 200 400 600 Time (days)

Figure 2 (b) Analysis of Uptake Data: Residuals vs. time. 256

I conclude that simple exponential equations quite adequately fit the uptake data, the loss data, and the steady-state concentration factors observed by Seymour'6). Because the single-compartment model fits the data so well, I also conclude that either the individualorgans of the oysters have closely similar turnover-rate constants for Znor at least that the dominant organs in whole-body Zn content turnover at similar rates. These results are especially significant because few turnover studies combine the quantity of data and duration in time as that above.

The Problem of Multiple Steps Where radionuclide uptake proceeds through multiple steps which have roughly similar first-order rate constants, uptake data, particularly during early times, will not follow the simple relationship used above. Aten(16) showed thata two-trophic-level system produced radionuclide concentration at the second level which initially increased in proportion to the second power of time. Such possibilities may have been obviated in Seymour's experiment because the oysterswere planted in Willapa Bay after the rest of the Bay ecosystem had ample timeto approach its own steady-state level.

The Problem of Multiple Compartments Where an element is present in an organism in several compartments (organs, types of binding sites, etc.), the whole-organism kineticsmay be more complicated than the above equations would indicate. To demonstrate this possible source of difficulty, I shallassume that the pro- tein-bound Zn which Wolfe(17) found inCrassostrea virginica(96% of total body Zn) turns over with an effective half-life of 135 days (A + S = 5.13 X 10-3 d-1) and the remaining 4% will beassumed to turn over with a 1-day effective half-life (A + S = 0.693 d-1). Calculated loss curves for different periods of uptakeare shown in Fig. 3. It can readily be seen that the loss of 6 Zn following 0.1 or 1 day ofup- take (lower two curves) is dominated by the smaller, rapid-turnover compartment; whereas, were uptake carried out for 135 dor 1000 d (upper two curves), the larger, slow-turnovercompartment would dominate. Experimental loss of 65Zn by mussels followscurves which are strikingly similar to Fig. 3(18).

It should be noted that radionuclide loss studies following short uptake periods are most relevant to simulation of accidentalspill conditions, that is, impulsesources. On the other hand, long uptake periods may be required in order to simulate chronic contamination situations. The latter is, of course, more relevant fordetermination of the role of organisms in trace element cycling.

Acknowledgement: This work was supported by the U.S. Atomic Energy Commission Contract AT (45-1) 2227, TaskAgreement 12, (RLO-2227-T12-36). I thank Drs. R. Holton, W. Renfro, L.Small, andH. Vanderploeg for many helpful discussions and suggestions. 257

100 Uptake Time /000 d /35d

I0 /Od

/Od

0.1 Old

1 1 1 0.011 1 0 5 10 15 20 Loss Time (days)

Figure 3:Calculated Loss Curves for a Two-Compartment Model after Various Uptake Periods. 258

REFERENCES

1.D. A. Wolfe, levels of stable Zn and 65Zn in Crassostrea virginica from North Carolina.J. Fish. Res. Bd. Canada 27:47-57 (1970).

2. H.Boroughs,W. A. Chipman, and T. R. Rice, Laboratory experiments on the uptake,accumulation, and loss of radionuclides by marine organisms.Pages 80-87IN: The Effects of Atomic Radiation on Oceanography and Fisheries, NAS-NRC Publication 551, National Aca- demy of Sciences-National ResearchCouncil, Washington, D.C. (1957).

3.W. A.Chipman,T. R. Rice and T. J.Price,Uptake and accumulation of radioactive zinc by marine plankton, fish, and shellfish.U.S. Fish. Wildlife Serv. Fish. Bull. 58:279-292 (1958).

4.D. G. Watson, J. J.Davis,andW.C.Hanson,Zinc-65 in marine organisms along the Oregon and Washingtoncoasts. Science 133: 1826-1828 (1961).

5.A. Seymour andG.B.Lewis,Radionuclides of Columbia River origin in marineorganisms, sediments,and water collected from the coastal and offshore waters of Washington and Oregon, 1961-1963. USAEC UWFL-86 (1964).

6.A.Seymour,Accumulation and loss of zinc-65 by oysters in a natural environment. Pages 605-619 IN:Disposal of Radioactive Wastes into Seas, Oceans, and Surface Waters,IAEA, Vienna (1966).

7.E. 0. Salo and W. L.Leet,The concentration of 65Zn by oysters maintained in the discharge canal of a nuclear power plant. Pages 363-371 IN: Proc. 2nd National Symp. on RadioecoZogy, Ann Arbor, Michigan, 15-17 May 1967, USAEC CONF 670503(1969).

8.M. G.Romeril,The uptake and distribution of 65Zn in oysters. Marine Biology9(4):347-354 (1971)0

9. G. G. Polikarpov,RadioecoZogy of Aquatic Organisms [English trans.]. Reinhold, New York (1966),

10. G.B.Barinov, Exchange of45Ca,137Cs,and144Cebetween an alga and seawater. Okeanologiya5(1):111-116 (1965);English trans. IN: Oceanography5(1):83-87 (1965)0

11. W. Feldt, Introductory paper. Pages 9-18 IN:Marine Radioecology, Proceedings of the Second ENEA Seminar,Hamburg,20-24 Sept. 1971 (1971). 2!__ 9

12. D. C. Reichle,P. B.Dunaway,and D. J.Nelson,Turnover and concentration of radionuclides in foodchains. Nuclear Safety11(1): 43-54(1970).

13.National Academy of Sciences-National ResearchCouncil,Disposal of Low-Level Radioactive Wastes into Pacific Coastal Waters. NAS- NRC Publication 985. Washington,D.C. (1962). 14.L. Krumholz and R. Foster, Accumulation and retention of radioactivity from fission products and other radiomaterials by fresh-water organisms.Pages 88-95 IN:The Effects of Atomic Radiation on Oceanography and Fisheries, NAS-NRC Publication 551, National Aca- demy of Sciences-National Research Council, Washington, D.C. (1957).

15.N. R. Draper and H. Smith, Applied Regression Analysis. Wiley and Sons, New York (1966).

16.A. H. W. Aten, Radioactivity in marineorganisms. Proc. 2nd UN Conf. Peaceful Uses Atomic Energy, Geneva15:414-418 (P/547)(1958).

17. D. A. Wolfe,Zinc enzymesinCrassostrea virginica.J. Fish. Res. Bd. Canada27:59-69 (1970).

18. A. W.vanWeers, Uptake and loss ofzinc-65 andcobalt-60 by the mussel MytiZus eduZis L.Symposium on the Interaction of Radioactive Contaminants with the Constituents of the Marine Environment,Seattle, Washington, July 10-19, 1972 [preprint] (1972). 260

RLO-2227-T12-39 Submitted toJournal of Chemical Education

LINEAR INSTRUMENT CALIBRATION WITH STATISTICAL APPLICATION

Ingvar L. Larsen Research Associate School of Oceanography Oregon State University Corvallis, Oregon 97331

and

Jerome J. Wagner Research Assistant School of Oceanography Oregon State University Corvallis, Oregon 97331

Quantitative determinations involving instrumentation are oftenaccom- plished by establishing a calibration curve.Readings from the particular instrument are plotted against known concentrations of standards and the concentration of the analyte in the sample evaluated by the underlying relationship of the calibration line.If the calibration is linear, or nearly linear such that a straight line providesa reasonable approximation, the concentration of the sample can be readily determined if it falls within the range of the standards.The equation for a straight line is given by:

Y = MX + I (1) where the Y value (instrument response) isa function of the concentration (X value); M is the slope of the calibration line; andI is the ordinate intercept (vertical axis).It may happen that the calibration is not linear. If this occurs, several options are available.The unknown concentration can be determined graphically, if sufficient reproducible data pointsare obtained from the appropriate standards.Curve fitting techniques, facilitated by use of a computer, may also be used.The application of transformations such as Beer's Law (1) to photometric absorption data, withina limited range, usually produces an adequate linear calibration line.Acton (2) discusses the use of transformations and some simple suggestionsare given in reference (3). After establishment of the calibration line the analyst proceedsto analyze his samples.Usually this is done by taking repetitive measurements upon a sample, converting these determinations to concentration units via the calibration line and then reportinga mean and error term based upon the repetitive measurements.However, as stressed by Linning and Mandel (4), this 261 replication error may not providea realistic estimate of the precision for themethod. Why? Because the overall precision in establishing the calibr..- tion line may be poorer than that indicated by thesample replication errcr. The scatter of the data points establishing the calibrationline provides a measure of the expected error in analyzing the sample and may be larger than thatindicated by sample replicationerror (4).The statistical uncertainit:y in establishing the calibration line must be considered.

STATISTICAL APPLICATION For a first approximation, a straight line drawn through the data points may be sufficient to evaluate the concentration in the sample.For quantitative analysis the application of linear leastsquares produces an unbiased calibration line.The best line which can be fitted is theone which minimized the sum of the squares of the vertical distance between the datapoints and the fitted line (5).This is referred to as the regression of Yon X, X being the independent variable (concentration) and Y, the dependentvariable (instrument response). The assumptions underlying linear regression require thatthe data means at each concentration lie on the line Y = MX + I with theerrors associated with the X values small in comparison to those associatedwith the Y values; these Y values are normally distributed with equality of varianceat each corresponding X value, and are statistically independent (2).

In order to facilitate writing computational formulas, the symbolsas given in Table I will be used to designate the required calculations(after 2)0 The slope for the equation ofa straight line as given by Eq.(1) is:

Slope, M = Sxx (2)

The intercept of the calibration line withthe verticalaxis, which represents a background signal contribution in the absence of the analyte, is given by:

Intercept, I = Y - MX (3)

An estimate of the scatter in fitting the straight line, referredto as the standard error of regression is obtained bysubtracting the fitted regression line, Y, from each data point, Y, summing the squareddeviations, dividing by the degrees of freedom (N- 2) and extracting the square root. Thus, the standard error of regression,Sr, is calculated by:

2 E (Y - Y)2 Y (Sx Sr = E Sxx (4) N - 2 VS yN - 2 262

Table 1. Symbols indicating designated computations.

2 2 (EX)2 NEX2- (EX) Sxx = E (x - X) EX2 N N

(EY)2 NEY2-(EY)2 Syy = E(Y -Y)2 =EY2- = N N

Sxy = E (X - X) (Y - Y) EXY -EXEY NEXY - EXEY N N

Ex X = N

EY Y N

N = number of readings obtained

Using weighted values

(EWX) 2 EWEWX2-(EWX)2 Swxx = EW(X - Xw)2 =EWX2- EW EW

2 (EWY) EWEWY2- (EWY)2 Swyy = EW(Y - Yw)2 =EWY2- EW EW

Swxy = EW(X - Xw)(Y - Yw) EWXY -EWXEWY EWEWXY - EWXEWY EW EW

EWX Xw = EW

Yw = EWY EW W = appropriate weighting factor For statistical purposes computations using these formula should be made to more than the required significant number of figures with the final desired results rounded to the significant number of figures. 26

Manipulation of Eq.(1) allows the calculation of the amount of unknown in the sample, in terms of concentrationunits, to be made. After establishment of the calibration line,a sample instrument reading, Y', will have a corresponding concentration,

X' Y' - I M (5)

An error term, expressedas a confidence level for this value, X', is approximated (3, 4, 6) by: Confidence Level, C.L. =tMSr (XI - 1 +N+ SxxX)

where t is the Student t value for N-2 degrees of freedomand the desired confidence level.The other values are as previously defined.In some instances a mean value composed ofmore than one reading of the sample may be obtained, and in this case the value 1 under thesquare root sign becomes 1/Q, where Q represents the actual number of independentdeterminations performed on that particular sample (6, 7).An estimated confidence level for a mean of Q independent determinations is thus approximatedby:

2 C.L. =tSr 1 1+ X' X) M VQ N Sxx (7) where X' is the mean concentration correspondingto the mean instrument readings (Y') and evaluated similarto Eq. (5).Thus, a greater degree of confidence (resulting ina narrower confidence interval) can be placed on a mean of Q determinations than for a single observation. Tables II and III providean example of the calculations for estimating the confidence interval fora single sample observation and for a mean of 10 (Q = 10) independent measurementson the same sample.Table IV compares these resulting confidence intervals (expressedas a percent of the mean). As indicated, errors other than those associatedbetween sample readings must be considered:the calibration line scatter must be included. When there is a stronga priori knowledge that the relationship between the instrument response and concentration islinear, and the required regression assumpti?ns fulfilled, the confidence intervalcan be minimized by maximizing E(x - X): since the X values are assumed to be fixed, the choiceof the X values is often available (3).The value of E(X -3)2is maximized when the X observations are divided equally at thetwo extreme points of the desired range of the X values.Also, the narrowest confidence interval will result if the sample has a concentrationnear the means (X, Y) of the calibration line data. Examination of Figure 1 indicates the increase inthe confidence interval as values move along the calibration lineaway from the mean values, X and Y. 264

Table II. Example of data and calculations.

Standards & Instrument Calculated Values Blanks Response

(X values) (Y values) N = 15, Student t 2.160 (N- 2,0.95)= arbitrary units EX2= 22.50

0 0.019 EX = 15.00

0 0.024 (EX)2= 225.00

0 0.021 X = 1.00

0.5 0.498 EXY = 22.4575

0.5 0.521 EXEY = 225.8400

0.5 0.511 EY2= 22.419366

1.0 0.980 EY = 15.0560

1.0 1.014 (EY)2= 226.683136

1.0 1.002 Y = 1.003733

1.5 1.482 Sxx = 7.500

1.5 1.498 Syy = 7.307157

1.5 1.491 Sxy = 7.401500

2.0 1.972 Slope (Eq. 2), M = 0.987

2.0 2.025 Intercept (Eq. 3),I = 0.0169

2.0 1.998 Standard error of regression (Eq.4), Sr = 0.0148 Sample,Y' - 0.770 Sample concentration (Eq. 5), X' = 0.763

95% confidence level for one observation, (Eq. 6), C.L. = ± 0.0337

% C.L. _ (0.0337)(100) 0.763 4.42% 26-5

Table III. Example of data and calculatedvalues for replicate readings upon the same sample using calibrationline data given in Table II.

Independent Corresponding Sample Concentration Readings, Y' (Eq. 5), X'

0.770 0.763

0.778 0.771

0.761 0.754

0.765 0.758

0.773 0.766

0.777 0.770

0.772 0.765

0.768 0.761

0.773 0.766

0.763 0.756

Y' = 0.770 X' = 0.763

-XP - 10_3 Standard deviation, S.D.= xl = 5.72 X N

S.D. Standard error, S.E. = 1.81 X 10-3 V10 95% C.L. = (S.E.) (t) _ (1.81 X 10-3)(2.262)= 4.09 X10-3

%C. L. =(C.L.)(100) (4.09 X 10-3)(100) = 0.536 X' 0.763

95% confidence level forQ (Q = 10) observations (Eq. 7), C.L. = ± 0.0136

% C.L. =(0.0136)(100) 0.763 ±1.78 266

Table IV. Comparison of the 95% confidence levels(expressed as a percent of the mean) obtained by different procedures.

Replication Using calibration line

within sample (N = 10) Q = 1 Q = 10

0.536% 4.42% 1.78% 26 7

.e 0 4 a) C o,3 to CaD 2

0 I 2 3 4 5 Concentration (arbitrary units)

Figure 1. CALIBRATION LINE WITH CONFIDENCE INTERVALS 268

Wider confidence intervals result when extrapolatingbeyond the range of the data. For this reason the rule of thumb "it is safer to interpolate thanto extrapolate" applies to calibration line evaluations.

CALIBRATION LINES USING AVERAGE VALUES

Occasionally an excessive amount of calibration datamay be obtained and the analyst decides to use average instrument valuesto construct the calibration line rather than using total data. This may facilitate the handling of the data, but results in a loss in the degreesof freedom. It is the averages of the data points which then establish thescatter about the calibration line rather than the total individual points. Since the number of averages is less than the total number of data points, this will result inthe selection of a larger Student t valueand causesan increase in the confidence interval.

In using averages the appropriate weighting factors must be appliedto properly establish the calibration line. These weighting factors are generally the number of data points used in computing theaverage for the particular instrumental readings (2).

Table V gives numerical values obtained fora calibration line using the total data available and Table VI presents computations forthe same calibration line data based on using weightedmeans. A comparison of the resulting data is presented in Table VII. The values obtained using weightedmeans areessentially those obtained by using the total data (slight discrepanciesexist due to round- ing of data) with the exception of the confidenceinterval which depends upon the appropriate Student + value.

CALIBRATION LINE THROUGH THE ORIGIN

Occasionally the analyst may decide that the calibration line shouldpass through the origin of his axes; i.e., witha zero intercept. The equation for a line through the origin is given by:

Y = MX , (8) where the slope, EXY M= EN (9)

and an estimate of the variability, S, aboutthe line is given by:

EY2- (EXY)2 EE XT- (10) S V N - 1 269 Table V. Calibration line data with calculatedvalues usingtotal data.

Concentration Instrument Calculated Values Response

(X values) (Y values) Abitrary units

0 0.019 N = 21, t (N - 2, 0.95)= 2.093 0 0.024 EX2= 24.00 0 0.021 EX = 17.00

0 0.023 (EX) 2 = 289.00

0 0.020 X = 0.809524

0 0.021 EY2= 23.959155

0.5 0.498 EY = 17.1530 0.5 0.521 (EY)2= 294.225409 0.5 0.511 Y = 0.816810

0.5 0.513 EXY = 23.97650

0.5 0.515 EXEY = 291.601

1.0 0.980 Sxx = 10.238095 1.0 1.014 Syy = 9.94821

1.0 1.002 Sxy = 10.090738 1.0 1.005 Slope (Eq. 2), M = 0.9856 1.5 1.498 Intercept (Eq. 3), I = 0.0189 1.5 1.491 Standard error of regression (Eq.4), Sr = 0.0124 1.5 1.482 Sample concentration (Eq. 5), X' =0.833 2.0 1.972 95% confidence level (Eq. 6),C.L. = ±0.0269 2.0 2.025

2.0 1.998

sample,Y' =0.840 270

Table VI. Mean calibration line data and calculations.

Mean Mean Concen- Instrument tration Response w W X WX2 W Y WY2 W XY

X Y

0.0 0.021333 6 0.0 0.00 0.127998 0.002731 0.000000

0.5 0.511600 5 2.5 1.25 2.558000 1.308673 1.279000

1.0 1.000250 4 4.0 4.00 4.001000 4.002000 4.001000

1.5 1.490333 3 4.5 6.75 4.470999 6.663277 6.706498

2.0 1.998333 3 6.0 12.00 5.994999 11.980004 11.989998

E =21 17.0 24.00 17.152996 23.956685 23.976496

Number of means, N= 5. t = 3.182 (N-2,0.95)

Xw = 0.809524

Yw = 0.816809

Swxx = 10.238095

Swyy = 9.945958

Swxy = 10.090737

Slope, M = Swxy/Swxx=0.9856 Intercept, I = Yw- MXw = 0.0189 (Sway)2 Swyy Swxx Standard Error of regression, Sr= = 0.0124 N-2 Sample concentration (Y' = 0.840),X' = 0.833

95% confidence level, C.L. = tSr 1 1 +(X' -Xw)2= ± 0.0438 M V N Swxx 271

Table VII. Summary of data obtained from the two calibrationlines.

From total data From weighted means

Sample concentration, I 0.833 0.833

Degrees of freedom 19.000 3.000

Student t (N-2,0.95) 2.093 3.182

Standard error of regression 0.0124 0.0124

95% confidence level 0.0269 0.0438 272

The concentration of the sample can be obtained by manipulation of Eq. (8). For a samplereading,Y', the correspondingconcentration,X', is:

X' =-Y° M (11) and a confidence level estimate forX'is given approximately (3) by:

tS 1 + (X') 2 (12) C.L. = M ZX2 where t is selected for the appropriate confidence level with N- 1 degrees of freedom. The construction of this line may not actually pass through the mass center of the datapoints,as it is based upon the deviations of the data from the origin (2).Careful examination of the data should be made before adopting thismodel. In many instrumental analyses, signal contributions arising from blanks or instrumental noise (8) produce a background signal in the absence of the analyte which will result in a non-zero intercept for the calibration line.

Acknowledgments

Appreciation is expressed toDr.Norbert A. Hartmann of Oregon State University,Department ofStatistics,for reviewing the statistical concepts presentedherein. This work was supported by United States Atomic Energy Commission,Contract number AT(45-1)2227 Task Agreement 12 (RLO-2227-T12-39). 2"r3

Literature Cited

(1) Pinkerton,RichardC. J.Chem. Educ.,41, 336 (1964).

(2) Acton,Forman S. "Analysis of Straight LineData," Wiley,New York, 1959. Pp.219,20,11,86, 16.

(3) American Society for Testing andMaterials. Committee E-11 on quality control ofmaterials. Philadelphia,Pa,.,Spec,Tech. Publ.313, 1962, Pp. 13, 9, 7, 12.

(4) Linning,F. J. and John Mandel, Anal. Chem., 36(13),25A, (1964), (5) Grant,C. L.Statistics helps evaluate analytical methods,In: "Developments in AppliedSpectroscopy,"Vol.6, Proc,of the 18th annual mid-American spectroscopysymposium, Chicago,15-18 May 1967, W. K.Baer,A. J.Perkins,andE.L.Grove,ed,, PlenumPress,New York, N.Y., 1968. P. 122.

(6) Snedecor,George W. and William G.Cochran. "Statistical Methods," 6thed.,Iowa State UniversityPress, Ames, Iowa, 1968.P. 160.

(7) Mood,Alexander M. and Franklin A.Graybill, "Introduction to the Theory ofStatistics,"2nded.,McGraw-Hill Book Co.,Inc.,New York, 1963, Pp. 337-339.

(8) Coor, Thomas, J.Chem. Educ.45(7 &8),A533 (1968), 274

RLO-2227-T12-40

Accepted byDeep-Sea Research

Food Habits of Deep-Sea Macrourid Fishes off the Oregon Coast

William G. Pearcy School of Oceanography, Oregon State University, Corvallis, Oregon, USA

and Julie W. Ambler Biology Department, Linn-Benton Community College, Albany, Oregon, USA

ABSTRACT The abyssal macrourid fishes off Oregonare generalized feeders, as theory predicts for large searchers inan unproductive environment.Analy- sis of stomach contents revealed that fish feedon a wide range of foods, but the proportions of different taxa of foodvary among and within species indicating some food selectivity.Broad feeding overlaps occurred between large Coryphaenoides armatus andC.filifer,which fedmainly on squid and fish biomass, and between smallC.armatusand C.ZeptoZepis, which feed largely onepifaunalcrustaceans. Pelagic crustaceans were numerous in the stomachs of largeC. ZeptoZepis.Overlap among food of these macrourids was more than the overlap among theshallow-water,demersal fishes. Most food consisted of epifauna or pelagic animals.It is not known if thesquid,shrimp and fishprey,known to inhabit mesopelagic depths, have vertical distributions extending to 2800m where they were consumed, or whether macrourids forage far above thebottom. The fact that some squid beaks were from squids larger than the fish whichate them suggests that macrourids are sometimes scavengers and capitalizeon sinking carcasses.The importance of pelagic animals in the diet of large macrourids implies a direct trophic link between abyssobenthic and pelagic animals that may be important in the transport of energy and elements into the deep ocean. 27'

INTRODUCTION

The importance of the cod-likefishes of the family Macrouridae (grenadiers or rat-tails) in thedeep-sea has been recognized since the CHALLENGER Expedition (Gunther,1887). They are widespread geographically and are a rich group in numberof species: approximately 300 species,most of which are benthicor benthopelagic, although a few speciesare bathypelagic. Macrouridae are the dominantgroup of fishes, both in biomass and numbers, on the continentalslopes and adjoining abyssal plains ofmany regions of the world (Murrayand Hjort, 1912; Grey, 1956; Marshall, 1964, 1965; Kort, 1967; Iwamoto, 1970). The recent commercial exploitation of macrourids in the North Atlanticand North Pacific oceans attests to their abundance (Pechenik and Troyanovskii,1970; Novikov, 1970).

Studies using underwatercameras provide new evidence that macrourids are abundant near the bottom in slopeand abyssal waters. Sequential photographs over bait reveal that largenumbers of grenadiers are attracted "almost immediately", suggestingthat these fishes are motile and able to locate food rapidly in the deepocean (Isaacs, 1969; Dayton and Hessler, 1972). These photographs, as wellas those by Marshall and Bourne (1964), show that macrourids, because oftheir capacious gas-filled swimbladders are neutrally buoyant, swimmingor hovering over the bottom (Marshall,1965).

Macrourids are the dominant familyby weight and numbers in deepwater bottom trawl catches off Oregonand Washington. They are present on the upper continental slope and dominate the benthicfish catches on the lower slope from 1400-2800 mas well as on the abyssal plains from 2800 m to at least 3990 m and 834 km offshore(Day and Pearcy, 1968; Alton, 1972; Pearcy and Stein, unpublished).

Despite the recognized importanceof macrourids in the deep-sea benthic communities, little is known abouttheir food habits, particularly of abyssal species. Knowledge of their food habitswould be especially interesting because whether predatorsare specialists or generalists is relevant to the structure of the deep-sea benthoscommunity and because their food habits may provide insight intothe relative amounts of different types of food transported fromupper waters into the deep-sea food web.

Sanders'(1968,1969) time-stability hypothesispredicts that large numbers of stenotopic species evolveto form "biologically-accomodated communities" where physical conditionshave remained constant and uniform for long periods of time, suchas in the deep sea. If food is limited, abyssal animals such as macrouridsmay be expected to have specialized, stenophagus feeding habits. Dayton and Hessler (1972), on the other hand, contend that the maintenance of thehigh species diversity found for deep- sea benthic invertebrates is a result of biologicaldisturbance produced by the cropping of predators,reducing the possibility of resource competition and competitive exclusion. They emphasize the lack of food-niche specialization of deposit feedinginvertebrates and the generalized feeding habits of predators in the deepsea. For arguments against the Dayton- Hessler theory, see Grassle andSanders (1973). 2 76

The rapid appearance of macrourids at baitsuggests two rather independent food sources for the deepsea:(1) a "rain" of fine detritus which supports the deposit and filter feeding animals,and (2) widely separated falls of large food items which support animals like thegrena- diers (Isaacs, 1969).

If small particles are the predominantsource of energy for the deep- sea benthic community, macrourids would be expected to feed mainlyon small deposit feeders. If large, fast-falling particles (such as carcasses of

fishes, squids, shrimps, whales, etc.)are important (Isaacs, 1969 ; Dayton and Hessler, 1972), large motile animals like macrourids would be expected to utilize this food source. In this latter situation large fishes may not have an important role as predators of benthic invertebrates,and thus may not be an important "cropping disturbance". Actually, by feedingon corpses they may act to even out patchiness and unpredictability of scattered high organic inputs by distributing fecesover a large area (see Dayton and Hessler, 1972, and for a contrary view, Grassle and Sanders, 1973).

Whether macrourids feed mostlyon pelagic or benthic food, infauna or epifauna, is even in question. Marshall and Bourne (1964) observe that most grenadiers have longer anal than dorsal finrays which maintain the head close to the bottom. Further, the triangular snout and subterminal mouth is adaptive for "rooting in the ooze" for benthic food. However the presence, or even predominance, of pelagic animals reported in the stomachs of some species (Podrazhanskaya, 1967; Pechenick and Troyanoskii,1970; Okamura, 1970; Novikov, 1970; Haedrickand Henderson, 1974) raises some doubts about the importance of benthic animals in the food habitsof some Macrouridae.

METHODS Otter Trawl (6.7 foot rope) and beam trawl (2.7m wide) collections of macrourids were made on the continental slope and abyssal plain offOregon from 1961-1972.Mesh sizes of the trawl liners were 1.3 or 3.2cm (stretch). Most of the trawl collections are from Cascadia Abyssal Plain 120-270 km off central Oregon.Some are from Tufts Plain 565-834 km offshore and some were below 1000 m on the continental slope and abyssal plain off northern Oregon and off Washington.Fishes were preserved in formalin at sea after the body cavity was incised.In the laboratory, fishes were specifically identified.Identifications and nomenclatureare based on Iwamoto and Stein (in press) who include Nematonurus, Chalinura and other previouslyrecog- nized genera as subgenera within thegenus Coryphaenoides. The fullness of all non-everted stomachs was noted for mostspecimens. Animals in each stomach were identified to speciesor genus, when possible, and then sorted into taxonomic groups or unidentifable remains whichwere weighed (drained wet weight) to the nearest 0.1 g. Resultsare expressedin frequency of occurrence and wet weight. A total of 2295 fishes (including those with everted stomachs) were examined fromover 100 collections. Observations on whether stomachs were everted, empty or contained foodwere made on 2279 fishes. 2 7'7

Squid beaks were identified using dissectionsof identified specimens, Clarke (1962)and Akimushkin (1965).

RESULTS

Five species ofmacrourids,all in the genus Coryphaenoides, were collected off the Oregon and Washingtoncoasts.Coryphaenoides (Nematonuru) armatus (Hector,1875),C.(ChaZinuraZ ) ZeptoZepis (GUnther, 1978) and C. (Coryphaenoides) filifer (Gilbert,1895)were caught on abyssal plains (2700-3900 mdepth);they co-occur and are all commonly captured in the samecollection,i.e., they are syntopic (seeRivas,1964). Corphaenoides (Nematonurus) pectoralis (Gilbert,1892) andC. (Coryphaenoides) acroZepis (Bean, 1884)were captured from 1000-2200 m on the continental slope. The number of individuals examined of eachof different species of Coryphaeno- ides and the number with food intheir stomachs are summarized in Table 1. A large percentage of all species hadevertedstomachs. This necessitated examination of many individuals inan attempt to obtain an adequate sample of stomach contents.

Table 1.The number of macrourids examined and percentage with everted stomachs,not everted but emptystomachs,and stomachs with contents for 2279 specimens of the five species collected off Oregon.

Total # % everted% empty% with contents C.armatus 1044 73 0.2 27 C.fiZifer 837 92 1.0 7 C.ZeptoZepis 160 49 0.6 50 C.acroZepis 216 95 2.0 3 C.pectoraZis 22 55 27.0 18

Many different kinds of food itemswere found in the stomachs (Table 2), especially in the abyssal species: 84 in C.armatus, 41 in C. ZeptoZepis and 40 inC.fiZifer.The number of food taxa listed is obviously related to the number of fishes examined with recognizableprey in their stomachs and our skill ofidentification,but evenso,some qualitative differences in food habits areobvious. Many taxa ofamphipods,cephalopods, polychaetes and fishes were found in C. armatusstomachs,but of those only amphipod taxa were well representedin C.leptolepis and only cephalopods in C.filifer. Individual fish usually hada variety of food types in their guts, rather than many individuals of thesame food taxa.This was especially true for small fishes that fed moreon small invertebrates.On the average only 1.9 individuals per amphipod taxawere found in individual macrourid somachs.Of that 63 fishes that had amphipods in theirstomachs, most (60%) contained only a single amphipod.Both the diversity of food found in the species of macrourids and thevariety of food in individuals suggest that the three abyssal speciesare feeding generalists. 278

Table 2.Taxa identified from the stomach contents of macrourids.1

C. C. C. C. C. armatus leptolepis filifer pectoralis acrolepis

POLYCHAETES Maldanidae x Maldane sp. x Lumbrineridae x Lumbrineris x Onuphididae x x x Onuphus sp. x Nothriasp. x x Phyllodocidae Glyceridae Travesia sp. x x BIVALVIA Solemya agassizi x x Other bivalves x x x

CEPHALOPODA Gonatopsis borealis x x Gonatus magister x x Gonatus fabricii x x x Cranchiidae x x x x Onychoteuthidae x x Onychoteuthis banksi x Octopoteuthis sicula x x Unidentified squid beaks x x Japetella sp. x x x Vampyroteuthis infernalis x x

PYCNOGONIDA Colossendeidae x

CRUSTACEA Copepoda Phaennidae x x x Xanthocalanus sp. x x x Aetideidae x x x Scolecithricidae x x Euchaetidae k x x Mysidacea Small mysids x x x Gnathophausiai as x x Cumacea x 279

Table 2.(cont.)

C. C. C. C. C. armatus leptolepis filifer pectoralis acrolepis CRUSTACEA (cont.) Tanaidacea x

Isopoda Storthyngura sp. x x Fish-louse type x

Amphipoda x x x .x Gammaridea, Benth ic2 Aceroides edax x Bathymedon canididus x Bruzelia inlex x Epimeria sp. X Halice sp. Harpiniopsis triplex x Harpiniopsis excavata x x Harpiniopsis sp. X x Hippomedon tracatrix x Hippomedon sp. X Lepidepecreum sp. x Liljeborgia cota x x Monoculodes necopinus x Monoculodes recandesco x Oediceroides abyssorum x x Orchomene tabasco x x Schisturella totorami x Syrrhoe sp. x x Tryphosella metacaecula x Uristes sp. Uristes perspinis x Valettiopsis dentatus x

Gammaridea, Pelagic2 Lysianassidae X Eusirus sp. x Hyperiopsis sp. x Koroga megalops3 x x Paracallisoma coecum x x Paralicella sp. x x Paragissa sp. x X Pseudotiron sp.3 x x Rhachotropsis distincta x Rhachotropis sp. x Stegoce haloides sp. x x Synopiidae x Hyperiidea x 280

Table 2. (cont.)

C. C. C. C. C. armatus leptolepis filiferpectoralis acrolepis

Euphausiacea Thysanopoda sp. x x x Other euphausiids x

Decapoda Sergestes similis x x Bentheogennema n. sp. x x Pasiphaeidae x x Hymenodora acauthi- telsonis x Hymenodora glacialis x x x Systellapsis braueri x Crangon abyssorum x x x x Munidiopsis sp. x Chionecetes tanneri x megalops x Unknown crab x

HOLOTHURIOIDEA x x Protankyra pacificum x x

OPHIUROIDEA Ophiocten pacificum x x Opiocten sp. x Opiacantha sp. x x x hiura sp. x Scotoplanes sp. x

OSTEICHTHYES Tactostoma macropus x x Chauliodus macouni x Stomiatoid x Myctophidae x Searsiidae x -Macrouridae Vertebrae,. bones, flesh x x x x Fish scales(cycloid and ctenoid) x x x x

MISCELLANEOUS Plant Zostera x Fucus potato peel Nematoda x x x 281

Table 2. (cont.)

C. C. C. C. C. armatus leptolepis filiferpectoralis acrolepis MISCELLANEOUS (cont.) medusae x x Formaninifera x Echiurida x hooks-from unknown animals x x x x

No. Fish with stomach contents 290 80 60 5 11 lx indicates thelowest taxa to which a food item could beassigned. Hence genera listed do not necessarily belong in the family given immediately above.

2Distinctionbetween benthic and pelagic amphipods was basedon morphology (Barnard, 1962), Barnard (1969) and J. Dickinson(pers. comm.).

3Not known if pelagicor benthic. Table 3. Average frequency of occurrence and percent by weight of major food items in the stomachs of the three abyssal macrourids. Feeding overlap, CA, of Horn (1966), is given at the bottom of the table. CA for frequency of occurrence was calculated omitting the values for crustacean remains (in parenthesis) which were generally for fragments of crustaceans whose occurrencewas already represented. Trace amounts are indicated by a "T".

FREQUENCY OF OCCURRENCE PERCENT BY WEIGHT

C. C. C. C. C. C. armatus filifer leptolepis armatus filifer leptolepis

Crangon abyssorum 32 9 54 1.1 T 41.7 Storthyngura sp. 25 0 39 0.3 0 3.2 Amphipods 21 9 30 0.1 0.10 0.9 Copepods 43 3 48 0.1 0.03 T Shrimp 2 11 1 3.0 4.40 2.6 Mysids 3 11 10 0.6 8.80 2.8 Euphausiids 1 1 1 1.0 2.40 10.0 Unidentifieddecapods 5 11 9 1.5 1.10 4.4 Other crustaceans 6 0 1 0.4 6.10 0.0 Crustaceanremains (35) (43) (57) 3.8 1.60 27.1

Cephalopods 19 74 3 67.3 54.00 T Pelecypods-gastropods 3 2 1 0.1 0.05 0.4

Polychaetes 22 15 25 0.2 1.90 3.0

Ophiuroid 0 6 12 0.0 0.05 3.9 Holothuroid 16 20 0 7.6 3.80 0.0

Echiuroid 1 0 0

Fish 8 13 0 11.3 12.5 0.0 Fish scales 15 30 0 0.5 1.8 0.0 Fish or squid lenses 14 22 1 1.1 1.4 E 236 237 235 100.0 100.0 100.0

Overlap - C 0.22 J 1 0.54-'-' 0.96-'` 0.04 -'

0.86 1 ` 0.04 283

The average frequency of occurrence and percent of different fooditems by weight is given for each abyssal species in Table 3. Crustaceans, molluscs, polychaetes and enchinodermswerecommon to the diet of all three species. Based on occurrence ofprey items, benthic crustaceans, copepods, polychaetes and holothurians were frequent in C.armatusstomachs. Coryphaenoides fiZiferhad the lowest occurrence ofbenthiccrustaceans and the highest occurrence of cephalopods and fishes, whereasC. ZeptoZepis appears to be more of a benthicpredator feeding largelyon crustaceans like the shrimp Crangon abyssorumand the isopodStorthyngura. The larger proportion of benthic food in the stomachs of C.ZeptoZepisand the lower percentage of everted stomachs (Table1) suggest that this speciesmay have a smaller swimbladder and a higher specific gravity andthus it may have more demersal habitsthan the other macrourids.

Food Overlap

Horn's(1966)formula foroverlap,CA with possible range from 0.0 to 1.0 was used to estimate similarityamong the food of the three syntopic species of abyssalmacrourids. Feeding overlap basedon frequency of occurrence of different taxa indicatesa high degree ofsimilarity,0.86, between the diets of C. armatusand C. leptolepis.Both species had similar occurrence of manygroups,especially crustaceans and polychaetes. The overlap among the other speciesis less, 0.54 between C. armatus and C. fiZifer and 0.22 betweenC.filifer andC. similar diets. leptolepis, indicating less

On the basis of weight, cephalopods,mainly gonatid squids, dominated the diet of both C. armatusand C. fiZifer.For both of these species, fishes were next in importance,followed by holothuriansor crustaceans. Broad overlap of feeding is evidentbetween these species, C = 0.96, indicating high similarity of foodhabits. Crustaceans, on the other hand, were most important in the diet ofC. leptolepis; Crangon abyssorumwas the major food; it comprised 41.7% by weight of its stomachcontents. The amount o overlap among C.ZeptoZepisand either C.armatusand C.fiZifer is small (0.04) ona weight basis. The large difference between feeding overlap of C.armatusand C.ZeptoZepison a frequency of occurrence and weight basis (0.86 vs 0.04) is explained by the fact that althoughC. armatusfrequently ingests smallinvetebrates, such as crustaceans and polychaetes, these are relativelyunimportant on a total weight basis compared to squids and fishes whenallsizes of fishesare considered together. In general, these comparisonsindicate that these three grenadiers often feed on the same typesof prey, although the proportions of specific items sometimes vary significantlyamong species. Food Habits vs Size of Fish

The most apparent trend in the food habits of C.armatus(Table 4) is the decrease in the percentageof benthic crustacea (largelyCrangon abyssorumand Storthyngurasp.) with increasing size of fishand an increase in the percentage of cephalopodsand fishes by weight. A similar trend in the diet of C. armatusfrom Hudson Canyon in the AtlanticOcean 284

Table 4. Precent by weight that major identifiable food categories comprised of the diet for different sizes of three species of macrourids.

C. armatus Total Length in mm 100-299 300-399 400-499 500-599 >600 (No. Fish) (44) (44) (19) (4) (4) Benthic Crustacea 38 2 1 0 0 Pelagic Crustaceal 8 12 3 1 5 Crustacean remains 32 9 5 0 0 Cephalopods 1 45 60 94 1 Polychaetes 5 0 0 0 0 Holothurians 2 15 8 1 0 Ophuroids 0 0 0 0 0 Fishes 0 4 20 3 94 Miscellaneous 2 14 2 3 1 0 Percent Pelagic 9 61 83 97 100 Percent Unidentifiable3 66 35 17 5 8 C. fitifer Total Lengthin mm 200-399 400-499 500-599 600-699 (No. Fish) (3) (6) (17) (20) Benthic Crustacea 0 0 0 1 Pelagic Crustaceal 0 7 20 12 Crustacean remains 0 47 3 11 Cephalopods 33 1 64 47 Polychaetes 0 7 0 3 Holothurians 67 38 4 2 Ophuroids 0 0 0 0 Fishes 0 0 5 19 Miscellaneous 0 0 3 4 Percent Pelagic2 33 8 89 78 Percent Unidentifiable3 95 72 33 39 C. Zeptolepis Total Length in can 100-299 300-399 400-499 500-599 (No. Fish) (17) (18) (8) (2) Benthic Crustacea 59 56 44 28 Pelagic Crustaceal 0 5 4 45 Crustacean Remains 41 34 36 20 Cephalopods 0 0 0 0 Polychaetes 0 0 10 0 Holothurians 0 0 0 0 Ophuroids 0 3 3 7 Fishes 0 0 0 0 Miscellaneous 0 1 2 0 Percent Pelagic2 0 5 4 45 Percent Unidentifiable3 51 47 54 21

'includes identified mysids, euphausiidsand pelagic shrimps as well and orange sh ramp remains as red ifiable pelagic shrimps,cephalopodsand fishes 3Percent of total weight of stomach able. contents, identifiable and unidentifi- 285

was found by Haedrich and Henderson (1974). Food overlap (C ) on a weight basis is only 0.33 between 100-299 and 300-399 mm lengthclassesA based on the food categories of Table 3. Large fish eat large, pelagic prey and apparently compete little with the small fish that prey on benthic crustaceans.

Of the few smallC.fiZifer (200-500mm)that had non-everted stomachs, holothurians (largely Protankyra pacificum) and pelagic crustaceans comprised the most important portion of their stomachcontents,whereas gonatid squids and fishes increased in importance in largeindividuals,as they did withC. armatus. Coryphaenoides ZeptoZepis clearly depends on crustaceans for its food. Benthic crustacea are most important for all sizes, 100-500 mm in length, but a trend toward increasing proportions of pelagic crustacea (euphausiids and shrimps) and decreasing proportions of benthic crustacea is evident as fish size increases.Fishes and cephalopods were not important in the diet of any of the C. ZeptoZepis examined.Although the average food overlap based on weight between C. armatus and C. leptolepis is low (CA = 0.04, Table 3) the overlap among small individuals (100-299 mm) of these two species is high (CA = 0.85) because of the predominance of small benthic crustaceans in their stomachs.

DISCUSSION According to theories on feeding strategies, predators that spend much of their time in search rather than pursuit of prey should be feeding generalists, especially if the predator is large and inhabits an unproductive, non-patchy environment (Emlen, 1966; MacArthur and Pianka, 1966; Schoener, 1971; MacArthur, 1972).This conclusion accords with Ivlev (1961) and Hall et al.(1970) who observed that diets of fishes were more diversi- fied when food density was low.Deep-sea predators, therefore, should be generalists rather than specialists and consume a broad variety of foods. However actual categorization of the diet of an animal as "generalized" or "specialized" may not be easy.Quantitative criteria are seldom applied, and intuitive judgments may emphasize that the problem is largely one of semantics.For example, from the same data different authors reached opposite conclusions on whether some species of carnivorous deep-sea benthic invertebrates are food generalists or specialists (Solokova, 1959; Dayton and Hessler, 1972; Grassle and Sanders, 1973).Similarly the macrourids in our study could be called generalists or specialists depending on what data are selected on total number of prey taxa consumed, frequency of occurrence or weights of prey taxa, or overlap of food among the different species or sizes of fishes. In an attempt to answer the question of whether deep-sea fishes have more generalized diets than shallow-water fishes, we compared the amount of food overlap of the abyssal macrourids with that reported for northern, marine, demersal fishes by Tyler (1972).Reoccurrences of overlap of prey 2 86

types were calculated from ourdata in Table 3 using his methods.Princi- pal prey - food taxes thatoccurred in 10%or more of the individuals of a species or comprised 5% or more of the food weight-were listed in a predator-prey matrix tableand the number ofreoccurrences was computed as a percentage of the possible numberofreoccurrences. The percentage overlap among the three macrouridsbased on frequently occurringprey was 50%, compared to values of 15-24%given by Tylerfor fishes of Passamaquoddy Bay, N.S., and the IrishSea. Overlap based on prey comprising 5%or more of the weight was 12% for macrouridsand 10% for fishes from the Sea of Okhotsk.Although estimates of foodoverlap and specialization are influenced by the number and similarity of fishesand by the number and taxonomic breath of foodcategories,the Oregon rat-tails do notappear to be any more specializedor have any less food overlap than shallow-water fishes.Considering thesecomparisons,the high food overlap among food of some species and the diversifieddietsof all theabyssal macrourids (Tables 2 and 3), we conclude that they have relativelygeneralized food habits.This does not precludesome trophic specialization or prey selec- tionhowever. Indeed some selection isexpected. It is evidenced by the absence in the rat-tail dietsof common benthicinvertebrates,such as species of holothurians andotherechinoderms,and by differences in the feeding habits of differentspecies and sizes ofa species. Asgeneralists,smallmacrourids,which were primarily benthic feeders (Table 4), could be important in maintaining the high species diversityof the deep-sea benthos in accordance with the hypothesisof Dayton and Hessler(1972). Largemacrourids,on the other hand, fed mainly on pelagic animals and thus havea different trophic niche in the deep-seafood-web.

Other investigations havealso found that macrouridsare euryphagus and feed on a variety of both benthic and pelagic animals. Pechenick and Troyanovskii (1970) found thatMacrurus rupestrisin slope waters of the Northwest Atlantic feed intensivelyon zooplankton (shrimps, euphausiids CaZanusand amphipods), and they found seasonal bathymetric migrationsof M. rupestris to be relatedto changes in the vertical distributionsof these pelagicprey. The food habits of M. rupestrisfrom Icelandic waters were studies by Podrazhanskaya(1967)who found that the pelagic shrimp Pasipahea comprised 74.4%of thediet; amphipod, euphausiids and Themisto, a pelagic were also common.Novikov(1970)reported 34 taxa of food in stomachsof Coryphaenoides pectoraliscaught inslope watersof the North Pacific. Squids and shrimps were importantprey, although ctenophores, bryozans, echinoderms,amphipods,lan ternfishesand lipaaids sometimes were common (frequency of occurrence of10%ormore). The food habits of 25 species of macrourids caughton the continental slope that the off Japan revealed most important preywere euphausiids (Okamura, 1970).Euphausiids were found in the stomachs ofall but one species of grenadierand often constituted50-100%of the diet of individualspecies.Polychaetes, shrimps, fishes, squids and isopodswere next in overall ing the percentage of euphausiids importance,sometimes exceed- for individual species of macrourids. On the basis of food habitsOkamura concludes that are true benthic feeders, intimately some genera of macrourids associatedwith thesea floor, while others like the subgenus Nematonurusare primarily bathypelagic. 287

These papers indicate that pelagic animalsmay be more important than benthic animals as food of some species. We found that this was especially true for largeCoryphaenoides;over 50% of the fishes represented in Table 4 had diets consisting mostly of pelagic organisms. The occurrence of vertically migrating or mesopelagic animals suchas euphausiids, shrimps, squids and lanternfishes in the diet of slope-dwelling macrouridsis not unexpected, since these prey may migrate close to the bottom during the day (Pechenik and Troyanovskii, 1970). However, the presence of these animals in the stomachs of grenadiers takenon the abyssal plain off Oregon is indeed suprising. The squidsGonatus fabricii, G. magisterand Onychoteuthis banksii,the shrimpsHymenodora, Bentheogennema, Sergestes similisandSystellaspis braueri,euphausiids and the fishTactostoma macropusare all found within the upper 1000 m off Oregon, sometimes with peaks in abundance above 1000 m (Pearcy and Forss, 1966; Pearcyand Laurs, 1966). These animals are not known to be distributed to 2800m depth, the depth of the bottom where most macrouridswere collected, although few opening-closing net tows have been made below 1000m. There are several possible explanations of this perplexing but important problem of how these deep-sea fishes obtain their pelagic food: 1. After being caught in the trawl, macrourids fedon pelagic animals collected in the net from shallower depths. 2. The vertical distributions of pelagic preymay actually extend into bathypelagic waters close to the bottom. 3. Macrourids may undertake extensive feeding migrations intoover- lyingwaters. 4. The pelagic food items were actually consumedon or near the bottom after sinking as corpses through much of the water column.

A few pelagic animals from macrourid stomachs such as anOnycoteuthis banksii,a squid found at the surface at night (Clarke, 1966; Pearcy, 1965), anda Tactostoma macropus,which was caught in largest numbers above 500 m (Pearcy and Laurs, 1966), were in excellent condition and therefore recently ingested. Possibly this resulted from feeding in the trawls,as suspected for a pelagic shrimp by Judkins and Fleminger (1972). However, many of the other prey, like the gonatid squids, were often partially digestedor only identifiable from beaks and had obviously not been recently ingested. More- over, squids of the size eaten by macrourids are good avoiders and rarely if ever captured in our trawl nets. In fact, pelagic animals in general are rare in bottom trawl catches compared to benthic fauna; therefore, if indiscriminate net feeding occurred, benthic animals and sediment (which was rare) would be expected in the stomachs of macrourids.

Little is known about the vertical distribution of animals in deep water and possibly the ranges of some of the pelagic prey of macrourids in fact extend to near-bottom depths off Oregon. Small gonatid squids and Octopoteuthis sicula are commonwithin the upper 500 m (Pearcy, 1965; Degner, 1925), and G.fabriciiandG. magister are consideredto be usually mesopelagic (Clarke, 1966). However, Akimushkin (1965) believes that Gonatus fabriciiandGonatus magister,both important prey for C.armatus and C.fiZifer,are eurybathic ranging to depths of over 4 000 m. Clarke 288

and Merrett (1972) found cephalopod remainsin the stomachs of the demersal sharkCentroscymnus eoeZolepiscaught on long-lines in the North Atlantic at depths of 998-1975 m and believe thatsome species of squid live on or near the bottom at least some of the time. The ommastrephid squidIZZex has been commonly observednear the bottom at 1800 m on ALVIN dives in the North Atlantic (Hessler, personalcommunication). Macrourids may therefore feed on eurybathic squidsnear the vicinity of the abyssal ocean floor.

Pelagic shrimps found in the stomachshave also been caughtin mesopelagic waters, but again little isknown about their distributions below 1000 m. Sergestee simiZisis a diel migrator with a center of abundance usually above 500 m off Oregon. Hymendoraspp.,Bentheogennema n.sp. and Systellaspis braueri,also found in grenadier stomachs, have deeper centers of abundance than S.similis(Pearcy and Forss, 1966). Their distributions extend into bathypelagic waters below1000 m. BothS. braueriandH. gZaeialiehave been collected in closing midwatertrawls at depths of 2000- 2400 m off Oregon (Pearcy and Krygier,unpublished). According to Vinogradov (1970),Hymenodora gZacialisis distributed from 1000 m to 6000 m in the Kurile-Kamchatka Trench. He postulates a scheme of overlapping ontogenetic and diel vertical migrationsof pelagic animals to provide for active transport of organic matter fromsurface waters into the deep sea.

The best evidence for macrourids migratingvertically far above the bottom where they forage on pelagic animalspertains to some of the slope species. A few C.acrolepishave been collected in midwater trawls off Oregon a minimum of 1000 m above thebottom (Pearcy, unpubl.; Iwamoto and Stein, in press). Kort (1967) reports a similar finding for this species on the basis of stomach contents. He cites Birshtein and Vinogradov (1955) who consider C.aerolepsisto be a bathypelagic form, though they found that stomach contents also contained benthicisopods. SometimesM. rupestris rise 100-200 m above the bottomas indicated in echograms and low catches on the bottom (Pechenik and Troyanovskii, 1970). Okamura (1970) noted that C. pectoralisand C.Zongifilis(subgenusNematonurus)rely on squids and fishes for a great part of their foodand he believes this subgenus may swim well away from the sea floor andpossess bathy- or mesopelagic habits. No evidence exists to our knowledge, however,for extensive migrations of abyssal species into upper waters thatcould explain the occurrenceof mesopelagic prey, although Haedrich and Henderson(this issue) invole extensive vertical migrations of C.armatusto explain animals likeCliauliodusin the diet of this rat-tail.

Finally, animals from water abovethe bottom may die and sink to the bottom where they are devoured by fishes. Photography has revealed rapid localization and consumption of bait bymacrourids and other deep-sea fishes, suggesting that theyare indeed efficient searchers and scavangers (Isaacs, 1969; Curtis, 1971; Daytonand Hessler, 1972). Though we found no evidence for ingestion of pieces oflarge fishes or marine animals, as did Clarke and Merrett (1972), the relative sizeof squid beaks ingested suggests the possibility thatsometimesportions of dead cephalopods were eaten by grenadiers off Oregon. 289

The length of the lower rostrum ofbeaks from gonatid squids was re- gressed on body weight using intact specimensand data from Clarke (1962). The estimated median size of gonatidsquids in C.arrmatus stomachs was 100 g (range 1.4-1100 g) and inC. fiZifer was 205 g (range 8.2-3000 g). The majority of squids ingestedwere probably eatenwhole. However, the estimated weights of three squidswere actually larger than the fish that ate thebeaks,and the weights estimated from 6or 26 beaks were over one- half the weight of the fish fromwhich the beaks werefound. Since large squids are good swimmers and, ifhealthy, are probably too fast to be caught bymacrourids,dead or dying squids may sink to the bottomwhere they are eaten whole or in pieces bymacrourids. The possibility exists that other pelagic organisms found in grenadierstomachs may also have been eaten after they died and settledto thebottom. Because of the slow microbial decomposition of organic matterin the deep sea(Jannasch,Eimhjellen, Wirsen andFarmanfarmian, 1971),such corpses may constitute a fairly stable food supply for motile abyssobenthicscavangers. We believe that these macrouridsare both predators and scavangers, and that this is another indicationof their generalized foraging habits in the deep sea.Our data do not support the idea that theyare dependent on large organicfalls,. however,since small pelagic and benthic animalscom- prise much of the diet.Pelagic animals are especially important for large macrourids,but we are unable tosay where in the water column they are devoured.

ACKNOWLEDGEMENTS

We are grateful to Andrew Carey for providingmany of the fishes from his collections, to David Stein,John Dickinson, Danil Hancock and Robert Wasmer for identification, respectively,of macrourids, amphipods, polychaetes, and shrimps and to C.B. Miller for criticismof the manuscript. This research was supported by theU.S. Atomic Energy Commission (Contract No. AT(45-1)-2227, Task Agreement 12;RLO-2227-T12-40).

REFERENCES

Akimushkin, I.I. (1965) Cephalopods of the Sea of the U.S.S.R.Akademiya Nauk SSSR. Institut Okeanologii (Translation), 1-223. Alton, M.S. (1972) Characteristics ofthe demersal fish fauna inhabiting the outer continental shelf and slopeoff the northern Oregon coast. In:The Columbia River Estuary and AdjacentOcean Waters (ed. A.T. Pruter and D.L. Alverson)Univeristy of Washington Press, p. 583-634. Barnard, J.L. (1962) South Atlantic abyssal amphipods collectedby the R/V VEMA. In:Abyssal Crustacea by J.L. Barnard, R.J. Menziesand M.C. Bacescu.Columbia University Press, 1-78. 290

Barnard, J.L. (1969) The families and genera of marine gammaridean amphipoda. U.S. National Museum Bulletin 271, 535 p.

Birshtein, Ya.A. and M.E. Vinogradov (1955) Remarks on the food of deep- sea fishes in the Kurile-Kamchatka Trench. Zoologicheskii Zhurnal 34, 842-849.

Clarke, M.R. (1962) The identification of cephalopod "beaks" and the relationship between beak and size and total body weight. Bulletin of the British Museum (Natural History) Zoology 18 (10), 421-480.

Clarke, M.R. (1966) A review of the systematics and ecology of oceanic squids. Advances in Marine Biology 4, 91-300.

Clarke, M.R. and N. Merrett (1972) The significance of squid, whale and other remains from the stomachs of bottom-living deep-sea fish. Journal of the Marine Biological Asscoiation of the United Kingdon 52, 599-603.

Curtis, T.C. (1971) The role of large falls of organic material in the nutrition of thedeep-seamegafauna. M.S.Thesis. University of Califronia San Diego Library. 96 p.

Day, D.S. and W.G. Pearcy (1968) Species associations of benthic fishes on the continental shelf and slope off Oregon. Journal of the Fisheries Research Board of Canada 25, 2665-2675.

Dayton, P.K. and R.R. Hessler (1972) Role of biological disturbance in maintaining diversity in the deep-sea. Deep-Sea Research 19, 199-208.

Degner, E. (1925) Cephalopoda. Report on the Danish oceanographical Expeditions 1908-1910 to the Mediterranean and Adjacent Seas. Vol. II Biology, 1-94.

Emlen, J.M. (1966) The role of time and energy in food preference. The American Naturalist 100, 611-618.

Grassle, J.F. and H.L. Sanders (1973) Life histories and the role of disturbance. Deep-Sea Research 20, 643-659.

Grey, M. (1956) The distribution of fishes found below a depth of 2000 meters Fieldiana: Zoology 36(2),1-337.

GUnther, A.(1887) Report on the deep-sea fishes collected by H.M.S. CHALLENGER during the years 1873-76. Report on the Scientific Results of the Voyage of the H.M.S. CHALLENGER during the years 1873-76. Zoology 22, 1-335.

Haedrich, R.L. and N.R. Henderson. Pelagic feeding by Coryphaenoides armatus, a deep benthic rattail. Deep-Sea Research (this issue). 291

Hall,D.J., W.E.Cooper,and E.E. Werner (1970) An experimental approach to the production of dynamics and structure of freshwater animalcom- munities. Limnology and Oceanography 15, 839-928.

Horn, H.S. (1966) Measurementof "overlap" in comparative ecological studies. The American Naturalist 100, 419-424.

Isaacs, J.D. (1969) The nature of oceanic life. Scientific American 221 (3), 146-162.

Ivlev, V.S. (1961) Experimental ecology of the feeding of fishes. Yale University Press. 302 p.

Iwamoto, T. (1970) The R/V PILLSBURY Deep-Sea Biological Expedition to the Gulf of Guinea, 1964-1965. 19. Macrourid Fishes of the Gulf of Guinea. Studies in Tropical Oceanography 4 (part 2), 316-431.

Iwamoto, T. and D.L. Stein. A systematic review of the rattail fishes (Macrouridae: Gadiformes) from Oregon and adjacent waters. Bulletin of the California Academy of Sciences (in press).

Jannasch,H.W., K. Eimhjellen, C.O. Wirsen and A. Farmanfarmian (1971) Microbialdegratation of organic matter in the deep sea. Science 171, 672-675.

Judkins, D.C. and A. Fleminger (1972) Comparison of foregut contents of Sergestes similis obtained from net collections and albacore stomachs. Fishery Bulletin 70, 217-223.

Kort, V.C. (ed.) (1967) Biology of the Pacific Ocean. Fishes of the Open Waters. Whiptails (family Macrouridae or Coryphaenodidae),p. 221-322 (Translation U.S. Naval Oceanographic Office).

MacArthur, R.H. (1972) Geographical Ecology, Harper & Row.269 p.

MacArthur, R.H. and E.R. Pianka (1966) On optimal use of a patchy environment. The American Naturalist 100, 603-610.

Marshall, N.B. (1964) Bathypelagic macrourid fishes. Copeia 1964(1) 86-93.

Marshall, N.B. (1965) Systematic and biological studies of the Macrourid fishes (Anacanthini-teleostii). Deep-Sea Research 12, 299-322.

Marshall, N.B. and N.B. Bourne (1964) A photographic survey of benthic fishes in the Red Sea and Gulf of Aden, with observationson their population, density, diversity and habits. Bulletin of the Museum of Comparative Zoology, Harvard University 132 (2), 223-244.

Murray, J. and J. Hjort (1912) The Depths of the Ocean. Macmillian and Co. Ltd. London. 821 p. 292

Novikov, N.P. (1970) Biology of Chalinurapectoralis in the North Pacific. In: Societ Fisheries Investigations in the Northeastern Pacific,Part V. (P.A. Moiseev, ed). Pacific Scientific ResearchInstitute of Fish- eries and Oceanography (TINRO) 72, 304-331. (Israel Program for Scientific Translations).

Okamura, 0. (1970) Studies on the macrouridfishes of Japan. Reports of the Usa Marine BiologicalStation 17 (1-2), 1-79.

Pearcy, W.G. (1965) Species composition anddistribution of pelagic lopods from the Pacific cepha- Ocean off Oregon. Pacific Science 19, 261- 266.

Pearcy, W.G. and C.A. Forss (1966) Depth distribution of oceanic (Decapoda: Natantia) shrimps off Oregon. Journal of the Fisheries Research Board of Canada 23, 1135-1143.

Pearcy, W.G. and R.M.Laurs (1966) Vertical migration and distributionof mesopelagic fishes offOregon. Deep-Sea Research 13, 153-165.

Pechenik, L.N. andF.M. Troyanovskii (1970) Trawling resources on the North Atlantic continental slope. 65 pp. (Israel Program for Scien- tific Translations).

Podrazhanskaya, S.G. (1967) Feeding of Macrurusrup s i area. in the Iceland Annaly Biologii (Moscow)24, 197-198.

Rivas, L.R. (1964) A reinterpretation of the concepts "sympatric"and "allopatric" with proposal of the additional terms "syntopic"and "allotopic". Systematic Zoology13,42-43.

Sanders, H.L. (1968) Marine benthic diversity: a comparative study. The American Naturalist 102,243-282.

Sanders, H.L. (1969) Benthic marine diversityand the stability-time hypothesis. Brookhaven Symposia inBiology 22, 71-80.

Schoener, T.W. (1971) Theory of feedingstrategies. Annual Review of Ecology and Systematics2, 369-404.

Sokolova, M.N. (1959) The feeding ofsome carnivorous deep-sea benthic invertebrates of thefar Eastern Seas andthe Northwest Pacific Ocean. (In Russian) In: Marine Biology. B.N. Nikitin, editor. Transactions of theInstitute of Oceanology20, 227-244. tion byA.I.B.S.). (Transla- Tyler, A.V. (1972) Food resource division fishes. among northern, marine, demersal Journal of the FisheriesResearch Board of Canada 29, 997- 1003. - Vinogradov, M.E. (1970) Vertical distribution of the oceaniczooplankton. Israel Program for Scientific Translations.339 p. 293

Published in: OSU. Water Research Institute. 1973. Heavy Metals in the Environment:A Seminar.Corvallis, Oregon. pp. 37-44.

Presented October 5, 1972 by Norman H. Cutshall, School of Oceanography, Oregon State University.

HEAVY METALS IN ESTUARIES AND THE COASTAL ZONE

Metals, as a class, are among the substances man finds most useful. The history of civilization includes periods in time named after metals-- the Iron Age, the Bronze Age. Even the Atomic Age might have been called the "Uranium Age".

Each phase of metal processing results in release of metals to the environment. Mining exposes metal-rich rocks to accelerated weathering. Smelting and refining commonly result in the release of minor byproduct metals as waste of one form or another. Indeed the recovery or partial recovery of minor metals is often incidental to waste disposal. "Thallium is recovered from cadmium-containing flue dusts----". (Bateman 1950 p. 628) Selenium "is another minor by-product obtained from the refining of copper ores. ----Much more could be saved from flue dusts if the demand should justify it." (ibid. P. 622) "Some 60,000 to 70,000 tons of white arsenic are normally produced annually as a by-product of smelter smoke from arsenical ores. ----In most places the available material is greater than consumption. Sweden alone could produce enough for the entire world and has difficulty in getting rid of the poisonous material." (ibid. 609- 610)

In use, metals are often subject to corrosion and wear which leads to loss into the environment. Some usesinvolve direct release into the environment either intentional or unintentional. Application of lead arsenate insecticide or phenyl mercuric acetate fungicide introduces heavy metals to the environment. The addition of tetraethyl lead to gasoline and of chromium and zinc to waters used for industrial cooling leads to environmental contamination.

Whether released to air or water or soil, heavy metal contamination is eventually carried into estuarine and coastal water systems. Estuaries serve as funnels, through which land runoff is transported into the coastal ocean. In some cases estuaries appear to effectively trap or delay a good deal of the heavy metal flux passing through them (Turekian 1971). But estuaries are more than funnels or traps. Theyare also avenuesby which anadromous fishes travel from spawning beds to the sea and return. They are spawning and nursery grounds for a wide variety of coastalorganisms. Consequently, adverse changes in estuaries may be reflected in the biota of a broad geographic area beyond the estuary itself.

Figure 1 is a highly schematic representation of a coastal ecosystem. Heavy metals from mostsources areintroduced into either the water or sediment phases of the system. These phases normally exchange material 294

with one another and with the biotic phase in a dynamic partitioning. Added heavy metals are subject to thesameenvironmental partitioning. Concern over heavy metal contamination generally is focused upon effects in the biota. The magnitude and significance of such effects depends not only upon the nature of the source (quantity released, which metals are released, etc.) but also upon the details of how the contaminant is distributed within the ecosystem.

WATER

EFFECTS

SEDIMENT

Fig. 1. Idealized flow ofcontaminants through a coastalecosystem.

Heavy metals normally occur at very low concentrations in coastal waters (Table 1). Thus the addition of relatively small quantities of these elements to estuaries can potentially affect the concentrations. As an extreme example, the concentration of beryllium is only 6 x 10-8 grams/liter in seawater. This amounts to about half a cubic foot of the metal in the 200 mile long Chesapeake Bay, the largest estuary in the United States. 295

TABLE I

Concentrations of Selected Heavy Metals inSeawater(a)

Metal Concentration Vg/liter

Be 6 x 10-4 Cr 0.5 Cu 3 Zn 10 As 2.6 Se 9 x10-2 Ag 0.3 Cd 0.1 Hg 0.2 Pb 0.03 Bi 0.02

a. From Goldberg etal. (1971)

Low concentrations of heavy metals in solution are a resultof their tendency for sorption on particulate matter including suspended sediment, bedsediment,and small organisms.In somecases,active removal of metals from the water by organisms may also contribute to lowered concentrations in thewater. The transport and dilution of heavy metals in coastal systems is therefore affected by the movement of particulate matter as well as by the flow ofwater. The influence of initial sorption of metals upon their subsequent "availability" to organisms is one of the most significant problem areas in environmental research.How effective are sediments at de- toxifying contaminated waters? Under what conditions and to what degree will metals removed into sediments become incorporated into the biosphere? Mercury illustrates these questions well.Inorganic divalent mercury added to river waters rapidly becomes associated with particulate material (Bothner and Carpenter 1972).It is then susceptible to accumulation in bottom sediments as happened at Lake St. Clair (Chem. and Eng. News, 1970). Jernelov (1969) showed that mercury thus associated with sediments could be converted to methyl mercury (CH3Hg+) through microbial activity. 296

Mercury in this formaopear_sto be less tightly retained by sediment and more readily taken up byorganisms. Jernelov(1970)simulated the burial of inorganic mercury contaminated sediments with uncontaminated sediment and observed the accumulation of methyl mercury by fish maintained in overlyingwater. He found that the uptake of methyl mercury rapidly decreased with increasing depth of burial of the. contaminated sediment. The addition of benthic macrofauna (tubificids orclams)to the system, however,led to increased uptake of buried mercury. Apparently the benthic organisms accelerate the transfer of methyl mercury to overlying water thus increasing its availability to fish. Thus one might expect to find both the sediment and biota in the vicinity of a mercury source to contain elevated mercuryconcentrations. Klein and Goldberg(1970)found that the Hg content of sediments in the vicinity of the Hyperion sewage outfall in Southern California was higher than that of sediments away from the outfall.On the otherhand,sessile organisms from the outfall area were not significantly higher in mercury than those collected farther away (Young1971). The expectation from Jernelov's (1970) experiment has not been fulfilled. A number of possible reasons are evident:Perhaps the number of samples taken is not statisticallyadequate, Perhaps the local current patterns near the outfall carry mercury away from the samplesites. Perhaps the natural rates of dispersion at the site greatly exceed the rates of mercury methylation/release. Perhaps the wrong species weresampled. If the par- tition of mercury is to be adequately known so that its fate and effects in similar systems can be predicted, we must learn which of these hypotheses stand scientific test.

COLUMBIA RIVER SYSTEM

Another example may be found in the Columbia Riversystem. Here radio- active65Zn,formerly introduced into the river at Hanford, Washington, provides a tracer for Columbia River zinc (Osterberg 1962).This radioisotope becomes associated with particulate matter and can be found in shelf sediments off the coast of Washington (Cutshall etal.1971). It has also been measured in a wide variety of coastal marine organisms (Osterberg et al.1964). Carey and Cutshall(1972)compared the distribution and specific activity (65Zn/totalZn)in benthic organisms known to process sediments,with the distribution and specific activity of sediments.They concluded that sediment associated with 65Zn was not always theprimarpp source of 65Zn to sedimentprocessers. They suggested that supply of b5Zn to these organisms might follow a detrital food web pathway. These examples show the complexity to heavy metal cycling in coastal ecosystems. They also demonstrate that the role of natural partitioning in the effects of heavy metal contamination issubstantial. Experiments in which the lethality of given water concentrations of metals are measured, although relatively easy toconduct,are not ecologically adequate. 297

Our present understanding of the rates and mechanisms of heavy metal partitioning is only adequate to highlight our ignorance of such factors.If we are to become able to predict the effects of heavy metals incoastal systems,even to recognize the subtleeffects,this understanding must be improved.

REFERENCES

1.Bateman, Alan M. (1950) Economic MineralDeposits,2nd edition, Wiley New York.

2. Bothner,Michael H. and Roy Carpenter (1972) Sorption-Desorption Reactions of Mercury with Suspended Matter in the Columbia River, Symposium on the Interaction of Radioactive Contaminants with the Constituents of the MarineEnvironment, Seattle,Washington, July 10-14,1972,International Atomic Energy Agency.

3. Carey,A. G. and N. H. Cutshall (1972) Zinc-65 Specific Activities from Oregon and Washington (U.S.A.) Continental Shelf Sediments and Benthic InvertebrateFauna,Symposium on the Interaction of Radioactive Contaminants with the Constituents of the MarineEnvironment,Seattle, Washington,July10-14, 1972,International Atomic Energy Agency.

4.Chemical and Engineering News (1970) Mercury Stirs More Pollution Concern 48 (26): 36-37.

5. Cutshall,N. H., W. C.Renfro,D. W. Evans and V. G.Johnson,(1971) Zinc-65 in Oregon-Washington Continental Shelf Sediments, Third RadioecologySymposium,OakRidge, Tennessee,May 10-12, 1971.

6. Goldberg,E. D., W.S. Broecker,M. G. Gross and K. K. Turekian (1971) MarineChemistry,ch. 5 in:Radioactivity in the Marine Environment, NAS-NRC,Washington, D.C.

7. Jernelov,A. (1969) Conversion of MercuryCompounds,ch. 4 in: ChemicalFallout,MortonW.Miller and GeorgeG.Berg (eds.) Charles C.Thomas,Springfield Illinois.

8. Jernelov,A. (1970) Release of methyl mercury from sediments with layers containing inorganic mercury at differentdepths,Limnol. and Oceanogr.15: 958-60.

9. Klein,David H. and Edward D. Goldberg (1970) Mercury in the Marine Environment,Environ.Sci.Technol.,4(9):765-768.

10.Osterberg,C. L. (1962) Zn-65 Content of Salps andEuphausiids,Limnol. andOceanogr.7(4): 478-479. 298

11. Osterberg, C.L., W. G. Pearcy and Herbert Curl, Jr. (1964) Radio- activity and its relationship to oceanic Food Chains, Jour. Mar. Research, 22 (1): 2-12.

12.Turekian, K. K. (1971) Rivers,Tributaries andEstuaries,ch. 2 in: Impingement of Man on theOceans,D. W. Hood (ed.), Wiley, New York. 13. Young, David R. (1971)Mercury in the Environment, Southern California Coastal Water ResearchProject,LosAngeles,California. 299

PUBLICATIONS

This bibliography updates the one published in the 1971 AEC Progress Report and includes only publications resulting from AEC-sponsored research in the School of Oceanography, Oregon StateUniversity.1

RESEARCH PUBLICATIONS

Beasley, T.M., and C.L. Osterberg. 1969. Spontaneous deposition of lead on silver and nickel from organic and inorganic digests. Analytical Chem. 41:541-543. (RLO-1750-72)

Bella, D.A. and J.E. McCauley, 1971. Environmental considerations for estuarine dredging operations. Pp. 458-482 in Proceedings of WODCON World Dredging Conference, New Orleans, La. (RLO-2227-Tl2-22)

Brownell, C.A. and J.E. McCauley. 1971. Two new gregarine parasites from the spatangoid urchin Brisaster latifrons. Zool Anz. 186:191-197. (RLO-1750-59)

Butler, J.L. and W.G. Pearcy. 1972. Swimbladder morphology and specific gravity of myctophids off Oregon. J. Fish. Res. Bd. Canada 29:1145- 1150. (RLO-2227-Tl2-25)

Carder, K.L., R.D. Tomlinson, and G.F. Beardsley, Jr. 1972. A technique for the estimation of indices of refraction of marine phytoplankters. Limnology and Oceanography 17(6):833-839. (RLO-2227-T12-20)

Carey, A.G., Jr. 1965). Preliminary studies on animal-sediment interrelationships of the Central Oregon Coast. Trans. of the Symposium on Ocean Science and Engineering 1:100-110. (RLO-1750-2)

Carey, A.G., Jr. 1966. Studies on the ecology of benthic invertebrate fauna in the Northeast Pacific Ocean off Oregon, U.S.A. P. 36 in Abstracts of Papers, Proc. 11th Pac. Sci. Congr., Tokyo, Vol.IT. Sci. Counc. Japan, Tokyo.

Carey, A.G., Jr. 1969. Zinc-65 in echinoderms and sediments in the marine environment off Oregon. Pp. 380-388 in Symposium on Radioecology: Proc. 2d Nat. Symp., Ann Arbor, Michigan, May 15-17, 1967. (CONF, 670503). 774 p. (RLO-1750-20) l McCauley, J.E., ed. 1971. Ecological Studies of the Columbia River Estuary and Adjacent Pacific Ocean. ProgressReport to AEC, 1 July 1970 through 30 June 1971. Ref. 71-18. Dept. of Oceanography, Oregon State University, Corvallis, Oregon. (RLO-2227-T12-10) 300

Carey,A.G.,Jr. 1972.Ecological observations on the benthic invertebrates from the central Oregon continentalshelf. Chap.20 in Pruter, A.T. and D.L.Alverson,eds, The Columbia River Estuary and Adjacent Ocean Waters: BioenvironmentalStudies. U. of WashingtonPress,Seattle. 868p. (RL0-1750-58)

Carey,A.G.,Jr. 1972. Food sources ofsublittoral,bathyal and abyssal asteroids in the northeast PacificOcean. Ophelia 10:35-47. (RLO- 2227-T12-13)

Carey, A.G.,Jr. 1972. Zinc-65 in benthic invertebrates off the Oregon coast. Chap.33 inPruter,A.T. and D.L.Alverson,eds, The Columbia River Estuary and Adjacent OceanWaters: Bioenvironmental Studies. U. of WashingtonPress, Seattle. 868p. (RLO-1750-55)

Carey,A.G.,Jr. A comparison of benthic infaunal abundance on two abyssal plains in the northeast PacificOcean.Accepted byDeep-Sea Research. (RLO-1750-67)

Carey, A.G., Jr., and N.H.Cutshall. 1973.Zinc-65 Specific Activities from Oregon and Washington Continental Shelf sediments and Benthic InvertebrateFauna. Pp. 287-305 in International Atomic Energy Agency, Radioactive Contamination of the Marine Environment (IAEA-SM-158). IAEA, Vienna. (RLO-2227-T12-26)

Carey,A.G., Jr. and D.R.Hancock. 1965. An anchor-box dredge for deep- seasampling. Deep-SeaRes. 12(6):983-984.

Carey,A.G.,Jr.,andH. Heyamoto. 1972.Techniques and equipment for sampling benthicorganisms. Chap.18 inPruter,A.T. and D.L. Alverson, eds, Columbia River Estuary and Adjacent Ocean Waters:Bioenviron- mentalStudies. U. of WashingtonPress, Seattle.868p. (RLO-1750-57)

Carey,A.G.,Jr.,W.C.Pearcy,and C.L.Osterberg. 1966. Artificial radionuclides in marine organisms in the northeast Pacific Ocean off Oregon. Pp. 303-319 in Disposal of Radioactive Wastes into Seas, Oceans and SurfaceWaters: Proceedings ofSym, Vienna,16-20 May1966. Inter- national Atomic EnergyAgency,Vienna. 898p. (RLO-1750-3)

Carey,A.G.,Jr.,and R.R.Paul. 1968. A modification of the Smith-McIntyre grab for simultaneous collection of sediment and bottomwater. Limnol. Oceanogr. 13(3):545-549..(RLO-1750-24)

Coleman, L.R. and B.G. Nafpaktitis.1972.Dorsadenayaquinae,a new genus and speciesof myctophid fishfrom the easternNorth Pacific Ocean. Contributionsin Science225:1-11. (RLO-1750-56)

Cross,F.A.,J.M. Dean,and C.L.Osterberg, 1969. The effect of temperature, sediment and feeding upon the behavior of four radionuclides in a marine benthicamphipod. Pp. 450-461 in Symposium on Radioecology:Proc. 2d Nat. Symp., AnnArbor, Michigan,May15-17,1967 (CONF670503).774 p. (RLO-1750-18) 301

Cross, F.A., S.W. Fowler,J.M.Dean, L.F. Small and C.L. Osterberg. 1968. The distribution of 65Zn intissues oftwo smallmarine crustaceans determined by autoradiography. J.Fish. Res.Bd.Canada25(11):2461- 2466. (RLO-1750-71)

Cross, F.A. and L.F. Small. 1967. Copepod indicators of surface water movements off the Oregon Coast. Limn. and ocean. 12:60-72.

Curl, H.Jr.,N. Cutshall, and C. Osterberg. 1965. Uptake of chromium (III) by particles in seawater. Nature 205(4968):275-276.

Curl, H., Jr., and L.F. Small. 1965. Variations in photosynthetic assimilation ratios in natural marine phytoplankton communities. Limnol. and Oceanogr. (Supplement to Volume 10) November, R67-R73.

Cutshall, N.H. 1973. Heavy metals in estuaries and the coastal zone. Pp. 37-44 in Heavy Metals in the Environment: A Seminar. OSU Water Research Institute, Corvallis, Oregon. 203 p.

Cutshall, N.H. (in press) Turnover of Zinc-65 in oysters. Accepted by Health Physics. (RLO-2227-Tl2-36)

Cutshall, N., V. Johnson, and C. Osterberg. 1966, Chromium-51 in seawater: chemistry. Science 152:202-203. (RLO-1750-5)

Cutshall; N., and C. Osterberg. 1964. Radioactive particle in sediment from the Columbia River. Science. 144(3618):536-537.

Cutshall, N., W. C.Renfro, D.W. Evans andV.G. Johnson. 1973. Zinc-65 in Oregon-Washington continental shelf sediments. Pp. 694-702 in Nelson, D.J., ed., Radionuclides in Ecosystems: Proceedings of the Third National Symposium on Radioecology (CONF-710501), Oak Ridge National Laboratory and Ecological Society of America, Oak Ridge,Tennessee. 1268 pp. (RLO-2227-T12-4)

Day, D.S. and W.G. Pearcy. 1968. Species associations of benthic fishes on the continental shelf and slope off Oregon. J.Fish. Res. Bd. Canada 25(12):2665-2675. (RLO-1750-43)

Donaldson, H.A. and W.G. Pearcy. 1972. Sound-scattering layers in the northeastern Pacific. J. Fish. Res. Bd. Canada 29:1419-1423. (RLO- 1750-62)

Eagle, R.J. 1969. First records from the northeastern Pacific of the deep- sea ophidioid fish Parabassogigas grandis. J. Fish. Res. Bd. Canada 26:1680-1685. (RLO-1750-38)

Evans, D.W. and N.H. Cutshall. 1973. Effects of ocean water on the soluble- suspended distribution of Columbia River radionuclides. Pp. 125-140 in International Atomic Energy Agency, Radioactive Contamination of the Marine Environment (IAEA-SM-158). IAEA, Vienna. (RLO-2227-Tl2-28) 302

Forster, W.O. 1972. Radionuclide distribution in the Columbia River and adjacent Pacific shelf sediments. Chap. 27 in Pruter, A.T. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. U. of Washington Press, Seattle. 868 p. (RLO-1750-49)

Forster, W.O. 1972. Radioactive and stable nuclides in the Columbia River and adjacent northeast Pacific Ocean. Chap. 26 in Pruter, A.T. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. U. of Washington Press, Seattle. 868 p. (RLO-1750-60)

Fowler, S. and L.F. Small. 1967. Moulting of Euphausia pacifica as a possible mechanism for transport of zinc-65 in the sea. Int. J. Oceanol. Limnol. 1:237-245.

Fowler, S.W., L.F. Small and J.M. Dean. 1969. Metabolism of zinc-65 in euphausiids. Pp. 399-411 in Symposium on Radioecology: Proc. 2d Nat. Symp. (CONF 670503). 774 p.

Griggs, G.B., A.G. Carey, Jr., and L.D. Kulm. 1969. Deep-sea sedimentation and sediment fauna interaction in Cascadia Channel and on Cascadia Abyssal Plain. Deep-Sea Research. 16:157-170. (RLO-1750-26)

Gross, M.G., A.G. Carey, Jr., G.A. Fowler, and L.D. Kulm. 1972. Distribution of organic carbon in surface sediment, northeast Pacific Ocean. Chap. 12 in Pruter, A.L. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. U. of Washington Press, Seattle. 868 p. (RLO-1750-27)

Haertel, L., and C. Osterberg. 1967. Ecology of zooplankton, benthos, and fishes in the Columbia River estuary. Ecology 48(3):459-4720 (RLO- 1750-6)

Haertel, L.S., C.L. Osterberg, H. Curl, Jr. and P.K. Park. 1969. Nutrient and Plankton ecology of the Columbia River estuary. Ecology 50:962-978. (RLO-1750-77)

Hansen, P.J. and W.O. Forster. 1969. Measurementof Columbia River flow time fromHanford's reactorsto Astoria,Oregon---Summer1966. Water Resources Research5:1129-1131. (RLO-1750-47)

Holton, R.L., W.O. Forster, C.L. Osterberg. 1973. The effects of gamma irradiation on the maintenance of population size in brine shrimp, Artemia. Pp. 1198-1205 in Nelson, D.J., ed, Radionuclides in Ecosystems: Proceedings of the Third National Symposium on Radioecology (CONF- 710501). Oak Ridge National Laboratory and The Ecological Society of America, Oak Ridge, Tennessee. 1268 pp. (RLO-1750-89) 303

Holton, R.L., W.O. Forster, C.L. Osterberg. 1973. The effect of gamma irradiation on the reproductive performance of Artemia as determined by individual pair matings. Pp. 1191-1197 in Nelson, D.J., ed, Radio- nuclides in Ecosystems: Proceedings of the Third National Symposium on Radioecology (CONF-710501). Oak Ridge National Laboratory and Ecological Society of America, Oak Ridge, Tennessee. 1268 pp. (RLO- 1750-88)

Hubbard, L.T., Jr., and W.G. Pearcy. 1971. Geographic distribution and relative abundance of salpidae off the Oregon Coast. J. Fish. Res. Bd. Canada 28:1831-1836. (RLO-1750-64)

Iwamoto, T. and D.L. Stein. (in press) A systematic review of the rattail fishes (Macrouridae:Gadiformes) from Oregon and adjacent waters. Occ. Pap. Calif. Acad. Sci. (RLO-2227-T12-35)

Jennings, D., N. Cutshall, and C. Osterberg. 1965. Detection of gamma- ray emission in sediments in situ. Science 148(3672):948-950.

Jennings, C.D., and C.L. Osterberg. 1969. Sediment radioactivity in the Columbia River Estuary. Pp. 300-306 in Symposium on Radioecology: Proc. 2d Nat. Symp. (CONF-670503). 774 p. (RLO-1750-21)

Jennings, C.D. and C.L. Osterberg. 1973.Specific activity of Iron-55 in Pacific Salmon. Pp. 703-708 in Nelson, D.J., ed, Radionuclides in Ecosystems: Proceedings of the Third National Symposium on Radioecology (CONF-710501). Oak Ridge National Laboratory and the Ecological Society of America, Oak Ridge, Tennessee. 1268 pp. (RLO-1750-86)

Johnson,V., N.H.Cutshall,and C.L.Osterberg. 1967. Retention of zinc- 65 byColumbia River water.Water Resources Res. 3(l):99-102. (RLO- 1750-7)

Kujala, N.F., I.L. Larsen, and C.L. Osterberg. 1969. Radioisotopemeasure- mentsin Pacific Salmon. Pp. 440-449 in Symposium on Radioecology:' Proc. 2d Nat. Symp. (CONF. 670503) 774 p. (RLO-1750-14)

Kyte, M.A. Recent Ophiuroidea occurring off Oregon, U.S.A. Submitted to J. Fis. Res. Bd. Canada. (RLO-2227-T12-12)

Kyte, M.A. A redescription of Ophiacantha rhachophora Clark and a description of sp. n. (Echinodermata: Ophiuroidea) Submitted to J. Fis. Res. Bd. Canada. (RLO-2227-Tl2-11)

Larsen, I.L., N.A. Hartmann, and J. J. Wagner. 1973. Estimating precision for the method of standard additions. Analytical Chemistry 45(8):1511- 1513. (RLO-2227-T12-23) 304

Larsen, I.L., W.C. Renfro, and N.H. Cutshall. 1973. Zinc-65 specific activity in Mytilus californianus tissues.Pp. 747-751 in Nelson, D. J, ed, Radionuclides in Ecosystems: Proceedings of the Third National Symposium on Radioecology (CONF-710501). Oak Ridge National Laboratory and The Ecological Society of America, Oak Ridge, Tennessee. 1268 pp. (RLO-2227-T12-5)

Larsen, I.L. and J.J. Wagner. 1973. Linear instrument calibration with statistical application. Submitted to Journal of Chemical Education. (RLO-2227-Tl2-39)

Laurs, R.M. and J.E. McCauley. 1964. A new acanthocephalan from the Pacific Saury. J. Parasitol. 50:569-571.

Loeffel,R.E. and W.O.Forster. 1970. Determination of movement and identityof stocks of Coho salmon in the oceanusing theradionuclide zinc-65.Research Reports of the Fish Commissionof Oregon 2:1-13. (RLO-1750-87)

McCauley, J.E. 1967. Status of the Heart Urchin,Brisasterlatifrons. J. Fish.Res.Bd. Canada. 24:1377-1384. (RLO-1750-11)

McCauley, J.E. 1972. A preliminary checklist of selected groups of invertebrates from otter-trawl and dredge collections off Oregon. Chap. 19 in Pruter, A.L. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. U. of Washing- ton Press, Seattle. 868 p. (RLO-1750-39)

McCauley, J.E. and A.G. Carey, Jr. 1967. Echinoidea of Oregon.J0 Fish. Res. Bd. Canada. 24(6):1385-1401. (RLO-2227-Tl2-12)

McCauley, J.E., M.A. Kyte, D.L. Stein and A.G. Carey, Jr. Biological fouling on TOTEM-1,a deep-sea sparbuoy. (Submitted for publication to Fish.Res.Bd.ofCanada) (RLO-2227-T12-8)

McCormick, J.M. 1965. Some aspects of the ecology of hydroids off the Oregon Coast. Northwest Science 39:139-147.

Neal, V.T. 1972. Physical aspects of the Columbia River and its estuary. Chap.2 inPruter,A.L. and D.L.Alverson,eds, The Columbia River Estuary and Adjacent Ocean Waters:Bioenvironmental Studies.U. of Washington Press,Seattle. 868p. (RLO-1750-40)

Osterberg, C. 1962.Fallout radionuclides in euphausiids.Science, 138 (3539):529-530.

Osterberg, C.1962. Zinc-65 in salps and euphausiids, Limnol. Oceanog., 7:478.

Osterberg, C. 1963. The effluent of the Columbia River as a source of radioactivity (abstract).Trans. Am. Geophys. Union 44:212-223. 305

Osterberg,C. 1968. Radioactivity from the ColumbiaRiver. Oc. Sci. and Oc.Engr. Symposium, 2:968-979. (RLO-1750-1)

Osterberg,C., A.G.Carey,Jr., andH. Curl,Jr. 1963. Acceleration of sinking rates of radionuclides in theocean. Nature 200:1276-1277.

Osterberg,C.L.,A.G. Carey,Jr., and W.G.Pearcy. 1966. Artificial radionuclides in marine organisms in the Northeast Pacific ocean off Oregon, U.S.A. Pp.275-276 in Abstracts ofPapers,No.320, Symp.IV. Second International OceanographicCongress,Moscow, USSR.

Osterberg,C., N.Cutshall,andJ. Cronin. 1965. Chromium-51 as a radioactive tracer of Columbia River atsea. Science 150(3703):1585-1587.

Osterberg,C.L., N.Cutshall,V.Johnson,J.Cronin,D. Jennings, and L. Frederick. 1966. Some non-biological aspects of Columbia River radioactivity. Pp. 321-335 in Disposal of Radioactivity Wastes into Seas, Oceans and Surface Waters:Proceedings ofSym.,Vienna, 16-20 May 1966. International Atomic EnergyAgency, Vienna. (RLO-1750-4)

Osterberg,C., L.D.Kulm,and J.V.Byrne. 1963. Gamma emitters in marine sediments near the ColumbiaRiver. Science. 139:916-917.

Osterberg,C., J. Pattullo andW. Pearcy. 1964. Zinc-65 in euphausiids as related to Columbia River water off the Oregoncoast. Limnology andOceanography. 9(2):249-257.

Osterberg,C., W.G.Pearcy,andH. Curl, Jr. 1964.Radioactivity and its relationshipto the oceanic foodchains. J.Mar.Res.22(1):1-12.

Osterberg,C.L.,W.G.Pearcy and N.Kujala. 1964. Gamma emitters in a fin whale. Nature. 204:1006-1007.

Osterberg,C.L. and R.W.Perkins. 1972. Methods for the measurement of Hanford-induced gamma radioactivity in theocean. Chap.25 in Pruter, A.L. and D.L.Alverson,eds, The Columbia River Estuary and Adjacent OceanWaters: BioenvironmentalStudies. U. of Washington Press, Seattle. 868p.(RLO-1750-70)

Osterberg,C., L.Small,andL. Hubbard. 1963. Radioactivity in large marine plankton as a function of surface area.Nature 197(4870):883-884.

Park,P.K.,C.L.Osterberg, W.O.Forster. 1972. Chemical budget of the ColumbiaRiver. Chap.5 inPruter,A.L. and D.L.Alverson,eds, The Columbia River Estuary and Adjacent Ocean Waters:Bioenvironmental Studies. U. of WashingtonPress, Seattle. 868p.(RLO-1750-41)

Pearcy,W.G. 1962. Egg masses and early developmental stages of the scorpaenidfish, Sebastolobus. J. Fish. Res.Bd. Canada,19:1169-1173. 306

Pearcy, W.G. 1964. Some distributional features of mesopelagic fishes off Oregon. J. Mar. Res. 22:83-102.

Pearcy, W.G. 1965. Distribution of oceanic cephalopods off Oregon, U.S.A. Proc. XVI Int. Congress Zoology 1:69.

Pearcy, W.G. 1965. Species composition, and distribution of pelagic cephalopods from the Pacific Ocean off Oregon. Pacific Science. 19:261-266.

Pearcy, W.G. 1970. Vertical migration of the oceanic shrimp Pandalas jordani, a feeding and dispersal mechanism. California Fish and Game 56:125-129. (RLO-1750-52)

Pearcy, W.G. 1972. Distribution and ecology of oceanic animals off Oregon. Chap. 17 in Pruter, A.L. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. U. of Washington Press, Seattle. 868p. (RLO-1750-45)

Pearcy, W.G. and J.W. Ambler. (in press) Food habits of deep-sea macrourid fishes off the Oregon coast. Accepted by Deep-Sea Research (RLO-2227- T12-40)

Pearcy, W.G. and R.R. Claeys. 1972. Zinc-65 and DDT residues in albacore tuna off Oregon in 1969. Calif. Mar. Res. Comm., CalCOFI Rept. 16: 66-73. (RLO-2227-Tl2-1)

Pearcy, W.G., and C.A. Forss. 1966. Depth distribution of oceanic shrimps (Decapoda; Natantia) off Oregon. J. Fish. Res. Bd. Canada 23(8):1135- 1143. (RLO-1750-16)

Pearcy, W.G., and C.A. Forss. 1969. The oceanic shrimp Sergestes similis off the Oregon coast. Limnol. and Oceanogr. 14:755-765. (RLO-1750-48)

Pearcy, W.G., and L. Hubbard. 1964. A modification of the Isaacs-Kidd midwater trawl for sampling at different depth intervals. Deep-Sea Res. 11:263-264.

Pearcy, W.G. and R.M. Laurs. 1966. Vertical migration and distribution of mesopelagic fishes off Oregon. Deep-Sea Res. 13:153-165. (RLO-1750-9)

Pearcy, W.G. and R.S. Mesecar. 1970. Scattering layers and vertical distribution of oceanic animals off Oregon. Pp. 381-394 in Farquhar, G. Brooke, ed, Proceedings of an International Symposium of Biological Sound Scattering in the Ocean. U.S. Govt. Printing Office, Washington D.C. (RLO-1750-65)

Pearcy, W.G., S. L. Meyer, 0. Munk. 1965. A "four-eyed" fish from the deep- sea: Bathylychnops exilis Cohen, 1958. Nature 207:1260-1262.

Pearcy, W.G. and C.L.Osterberg. 1964.Vertical distribution of radionuclides as measured in oceanicanimals. Nature 204:440-441. 307

Pearcy, W.G. and C.L. Osterberg. 1967. Depth, diel, seasonal and geographic variations in zinc-65 of mid-water animals off Oregon. Int. J. Oceanol. Limnol. 1(2):103-116. (RLO-1750-15)

Pearcy, W.G., and C.L. Osterberg. 1969. Zinc-65 and manganese-54 in albacore Thunnus alalunga, from the west coast of North America. Limnol. Oceanogr. 13(3):490-498. (RLO-1750-23)

Pearcy, W.G., G.H. Theilacker and R. Lasker. 1969. Oxygen consumption of Euphausia pacifica: The lack of a diel rhythm or light-dark effect with a comparison of experimental techniques. Limnol. Oceanogr. 14(2): 219-223.

Pearcy, W.G., andL.F. Small. 1968. Effects of pressure on the respiration of vertically migratingcrustaceans. J.Fish.Res.Bd.Canada 25(7): 13/1-/316.

Pearcy, W.G. and H.A. Vanderploeg. 1973. Radio-ecology of Benthic Fishes off Oregon. Pp. 245-261 in International Atomic Energy Agency, Radio- active Contamination of the Marine Environment. (IAEA-SM-158). IAEA, Vienna. (RLO-2227-Tl2-27)

Pearcy, W.G., and G.L. Voss. 1963. A new species of gonatid squid from the Northeastern Pacific. Proc. Biol. Soc. Wash. 76:105-112.

Pereyra, W.T., W.G. Pearcy, and F.E. Carvey, Jr. 1969. Sebastodes flavidus, a shelf rockfish feeding on mesopelagic fauna, with considerable of the ecological implications. J. Fish. Res. Bd. Canada 26:2211-2215. (RLO- 1750-44).

Renfro, W.C. 1972. Radioecology of Zinc-65 in Alder Slough, an arm of the Columbia River Estuary. Chap 29 in Pruter, A.L. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. U. of Washington Press, Seattle. 868 p. (RLO-1750-46)

Renfro, W.C. 1973. Seasonal radionuclide inventories in Alder Slough, an ecosystem in the Columbia River estuary. Pp. 738-746 in Nelson, D.J. ed, Radionuclides in Ecosystems: Proceedings of the Third National Symposium on Radioecology (CONF-710501). Oak Ridge National Laboratory and The Ecological Society of America, Oak Ridge, Tennessee. 1268 pp. (RLO-2227-Tl2-6)

Renfro, W.C., W.O. Forster, and C.L. Osterberg. 1972. Seasonal and areal distributions of radionuclides in the biota of the Columbia River estuary. Chap. 30 in Pruter, A.L. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies. U. of Washington Press, Seattle. 868 p. (RLO-1750-61)

Renfro, W.C., and C.L. Osterberg. 1969. Radiozinc decline in starry flounders after temporary shutdown of Hanford reactors. Pp. 372-379 in Symposium on Radioecology: Proc. 2d Nat. Symp. (CONF. 670503). 774 p. (RLO-1750-19) 308

Renfro, W.C.,R.V.Phelps, and T.F.Guthrie. 1969. Influence of preservation and storage on radiozinc and total zinc concentrations in fish. Limnol. Oceanogr. 14(1):168-179. (RLO-1750-25)

Renshaw, R.W. 1965. Distribution and morphology of the medusa, Calycopsis nematophora,from the North PacificOcean. J.Fish.Res.Bd.Canada 22(3):841-847.

Renshaw,R.W. and W.G.Pearcy. 1964.A new swivel cable clamp for towing large nets at differentdepths. Deep-Sea Res. 11:933-934.

Romberg,G.P. and W.C.Renfro. (1973). 32P specific activity and 65Zn concentrations used to study the movement and feeding behavior of young Chinooksalmon. J.Amer. Fish.Soc. 102(2):317-322. (RLO-1750-84)

Small,L.F. 1962. Use of radioisotopes in laboratory energy assimilation investigations withDaphnia, Nature,196:787-788.

Small, L.F. 1967. Energy flow in Euphausiapacifica. Nature. 215:515- 516.

Small,L.F. 1967.On the standardization of C14 for primary production estimates in aquaticenvironments. Iowa State J. ofSci.42(1):63-71.

Small,L.F. 1969. Experimental studies on the transfer of 65Zn in high concentrations byeuphausiids. J.Exp.Mar. Biol.Ecol.3:106-123.

Small,L.F. and F.A.Cross. 1972. Effects of the Columbia River plume on two copepodspecies. Chap.16 inPruter,A.L. and D.L. Alverson, eds, The Columbia River Estuary and Adjacent OceanWaters:Bioenvironmental Studies. U. of WhasingtonPress, Seattle. 868 p. (RLO-1750-45)

Small,L.F. and H.Curl,Jr. 1972. Effects of Columbia River discharge on Chlorophyll and light attenuation in the sea. Chap.9 in Pruter, A.L. and D.L.Alverson,eds, The Columbia River Estuary and Adjacent OceanWaters: EnvironmentalStudies. U. of WashingtonPress,Seattle. 868p.(RLO-1750-51)

Smiles,M.C. and W.G.Pearcy. 1971. Size structure and growth rate of Euphausia pacifica off Oregon.FisheryBull. 69:79-86. (RLO-1750-50)

Smoker,W.W.and W.G.Pearcy. 1970. Growth and reproduction of the lanternfish Stenobrachiusluncopsaurus. Jour.Fish.Res.Bd. Canada 27:1265-1275. (RLO-1750-63).

Stein, D. and J. Butler. 1971.A notacanthid Macdonaldia challengeri collected off the Oregon coast.J. Fish. Res.Bd. Canada 28:1349- 1350. (RLO-2227-Tl2-3)

Stein,D. and J. Butler. 1972. First records of Bathysaurus mollis (Pisces: Bathysauridae)from the Northeast Pacific Ocean.J.Fish.Res. Bd. Canada29:1093-1094. (RLO-2227-T12-24). 309

Sumich,J.L. and J.E.McCauley. 1973. Growth of a SeaUrchin,Allocentrotus fragilis,off the OregonCoast. Pacific Science 27(2):156-167. (RLO- 2227-T12-21)

Sumich, J.L. and J.E.McCauley. 1972. Calcium-magnesium ratios in the test plates of Allocentrotus fragilis.Mar.Chem. 1:55-59. (RLO-2227-T12- 29)

Tennant, D.A. and W.O.Forster. 1969. Seasonal variation and distribution of652n,54Mn,and 51Ct in tissues of the crab Cancer magister Dana. Health Physics 18:649-657. (RLO-1750-37)

Tomlinson,R.D. and W.C.Renfro. 1972, Losses of zinc-65 to inorganic surfaces in a marine algal nutrientmedium. Environmental Science and Technology 6(12):1001-1005. (RLO-2227-Tl2-9)

Vanderploeg,H.A. 1973. Rate of zinc uptake by Dover sole in the northeast Pacificocean: preliminary model andanalysis. Pp. 840-848 in Nelson, D.J., ed, Radionuclides inEcosystems: Proceedings of the Third National Symposium on Radioecology(CONF-710501). Oak Ridge National Laboratory and The Ecological Society ofAmerica,OakRidge,Tennessee. 1268 pp. (RLO-2227-Tl2-7)

Wasmer,Robert A. 1972.New records for four deep-sea shrimps from the northeasternPacific. Pacific Science26(3):259-263. (RLO-2227-T12- 2)

Wasmer, R.A. 1972. A new species of Hymenodora(Decapoda,Oplophoridae) from the northeasternPacific. Crustaceana22(1):87-91. (RLO-1750- 85)

Wasmer, Robert A.A redescription of Petalidium suspiriosum Burkenroad, 1937(Decapoda: Natantia). Submitted to Crustaceana. (RLO-2227-T12- 19) 310

THESES

Bertrand. G.R. 1970.A comparative study of the infauna of the central Oregon continental shelf.Ph.D. Thesis, OregonState University. 123 numbered leaves. (RLO-2227-T12-15).

Beasley, T.M. 1969. Lead-210 in selected marineorganisms. Ph.D. Thesis, Oregon StateUniversity. 81 numbered leaves. (RLO-1750-73).

Bergeron, D.J. 1967. Uptake and retention of zinc-65 from food of Euphausia pacificaHansen.Master's Thesis, Oregon State University. 60 numbered leaves.

Bolen, J.J. 1972. Seasonalconcentration,turnover and mode of accumulation of 32P by the juvenile starry flounder Platichthys stellatus (Pallas)in the Columbia RiverEstuary. Ph.D.Thesis,Oregon State University. 142 numbered leaves. (RLO-2227-T12-33). Brownell, C.L. 1970. The relationship between morphology and ecology in the spatangoid urchin Brisasterlatifrons. Master'sThesis,Oregon StateUniversity. 98 numbered leaves.

Buffo, L.K.1967. Extraction of zinc from seawater. Master's Thesis, Oregon StateUniversity. 49 numbered leaves.

Butler,J.L. 1971.Swimbladder morphology and buoyancy of northeastern Pacificmyctophids. Master's Thesis.Oregon State University. 35 numbered leaves. (RLO-2227-T12-16).

Cronin,J.T. II. 1967. Techniques of solvent extraction of organic material from natural waters. Ph.D.Thesis, Oregon State University. 79 numbered leaves.

Cross, F.A. 1967. Behavior of certain radionuclides in a marine benthic amphipod. Ph.D.Thesis,Oregon StateUniversity. 64 numbered leaves. (RLO-1750-32).

Cutshall, N.H. 1967. Chromium-51 in the Columbia River and adjacent Pacific Ocean. Ph.D.Thesis, Oregon StateUniversity. 64 numbered leaves.

Day, D.S. 1967. Distribution of benthic fishes on the continental shelf and slope off the Oregon coast.Master's Thesis, Oregon State Univer- sity.69 numbered leaves. (RLO-1750-30).

Donaldson,H.A. 1967. Sound scattering by marine organisms in the northeastern Pacific Ocean.Master'sThesis,Oregon State University. 75 numbered leaves. (RLO-1750-31). Evans, D.W. 1972. Effects of ocean water on the soluble-suspended distribution of Columbia Riverradionuclides. M.S.Thesis,Oregon State University. 57 numbered leaves. (RLO-2227-Tl2-38). 311

Frederick, L.C. 1967. Dispersion of the Columbia River Plume basedon radioactivity measurements. Ph.D. Thesis, Oregon State University. 134 numbered leaves.

Gile, J.D. 1972. Zinc, copper andmanganesein the razor clam, Siliqua atula. M.S. Thesis, Oregon State University. 52 numbered leaves. (RLO-2227-T12-34).

Haertel, L.S. 1969.Plankton and nutrientecology ofthe Columbia River Estuary. Ph.D.Thesis,Oregon StateUniversity. 71 numbered leaves. (RLO-1750-75).

Hancock, D.R. 1969. Bathyal and abyssal polychaetes (Annelids) from the central coast of Oregon. Master's Thesis, Oregon State University. 121 numbered leaves. (RLO-1750-53).

Hanson, P.J. 1967. Vertical distribution of radioactivity in the Columbia River Estuary. Master'sThesis, Oregon State University. 77 numbered leaves. (RLO-1750-80).

Hanson, P.J. 1969. Water tracing with soluble metal chelates and neutron activation analysis: a laboratory and field study. Ph.D. Thesis, Oregon State University. 100 numbered leaves. (RLO-1750-81).

Harney, P.J. 1973. Loss of zinc-65andmanganese-54 from the fresh water mollusc Anodonta. M.S.Thesis, Oregon State University. 107 numbered leaves. (RLO-2227-T12-42).

Holton, R.L. 1968. Effects of Co-60 irradiation on the reproductive performance of the brine shrimp Artemia salina. Ph.D. Thesis, Oregon State University. 74 numbered leaves. (RLO-1750-43).

Hubbard, L.T. 1967. Distribution and occurrence of Salpidae off the Oregon Coast. Master's Thesis, Oregon State University. 31 numbered leaves.

Hufford, G.L. 1967. Some aspects of the biology of the deep sea holothuroid Paelopatides sp. Master's Thesis, Oregon State University. 67 numbered leaves. (RLO-1750-29).

Jennings, C.D. 1966. Radioactivity of sediments in the Columbia River estuary. Master's Thesis, Oregon State University. 62 numbered leaves.

Jennings, C.D. 1967. Iron-55 in Pacific Ocean organisms. Ph.D. Thesis, Oregon State University. 81 numbered leaves. (RLO-1750-35).

Kujala, N.F. 1966. Artificial radionuclides in Pacific salmon. Master's Thesis, Oregon State University. 62 numbered leaves.

Larsen, I.L. 1970. Determination of Zn-65 specific activity in various tissues of the Californiaseamussel, Mytilus californianus. Master's Thesis, Oregon State University. 58 numbered leaves. (RLO-1750-78). 312

Laurs, R.M. 1967. Coastal upwelling and the ecology of lower trophic levels. Ph.D. Thesis, Oregon State University. 121 numbered leaves.

Lee, W.Y. 1971. The copepods in a collection from the southern coast of Oregon 1963. Master's Thesis, Oregon State University. 62 numbered leaves. (RLO-2227-Tl2-17).

Lenaers, W.M. 1972. Photochemical degradation of sediment organic matter: effect on zinc-65 release. M.S. Thesis, Oregon State University. 56 numbered leaves. (RLO-2227-T12-32).

Longaker, H.L. 1974. Growth response of a marine phytoplankton Coccolithus huxleyi,to various chemical forms of cobalt. M.S. Thesis, Oregon State University. 52 numbered leaves. (RLO-2227-Tl2-44).

McCormick, J.M. 1965. Some aspects of marine hydroid ecology off Oregon. Master's Thesis, Oregon State University. 88 numbered leaves.

McCrow, L.T. 1972. The ghost shrimp, Callianassa californiensis Dana, 1854, in Yaquina Bay, Oregon. M.S. Thesis, Oregon State University. 64 numbered leaves. (RLO-2227-Tl2-14).

Morrison, G.E. 1966. An investigation of the distribution of Ngphthys caecoides in Yaquina Bay. Master's Thesis, Oregon State University. 35 numbered leaves.

Osterberg, C.L. 1962. Radioactivityin oceanicorganisms. Ph.D. Thesis, Oregon State University. 125 numbered leaves. Pope, S. 1970. Antimony-124 in the lower Columbia River.Master's Thesis, Oregon State University. 57 numbered leaves. (RLO-1750-74).

Renfro, W.C. 1967. Radioecology of Zn-65 in an arm of the Columbia River estuary. Ph.D. Thesis, Oregon State University. 88 numbered leaves. (RLO-1750-36).

Romberg, G.P. 1969. Determination and application of P-32 specific activity in Columbia River fish. Master's Thesis, Oregon State University. 65 numbered leaves. (RLO-1750-82).

Smiles, M.C. 1969. Size, structure and growth rate of Euphausia pacifica off the Oregon coast. Master's Thesis. Oregon State University. 44 numbered leaves. (RLO-1750-79).

Squire,D.R. 1972. The Japaneseoyster drill,Ocenebrajaponica,in Netarts Bay,Oregon. M.S.Thesis,(GeneralScience),Oregon State University, 65 numberedleaves. (RLO-2227-T17-31). Stanford,H.M. 1971. The concentration and oxidation state of chromium in seawater. Master's Thesis, Oregon StateUniversity. 74 numbered leaves. 313

Sumich,J.L. 1970. Growth of a sea urchin Alloncentrotus fragilisat different depths off the Oregoncoast.Master's Thesis, Oregon State University. 51 numbered leaves. (RLO-1750-68).

Tennant, D.A. 1968. Distribution of Zn-65, Mn-54 and Cr-51 in the tissues of the Dungenesscrab,Cancermagister. Master'sThesis,Oregon State University. 51 numbered leaves. (RLO-1750-33).

Tipper,R.C. 1968. Ecological aspects of two wood-boringmolluscs from the continental terrace off Oregon.Ph.D. Thesis, Oregon State University. 137 numbered leaves.

Tomlinson,R. 1970. Non-biological uptake of zinc-65 from a marine algal nutrient medium.Master's Thesis, Oregon StateUniversity. 62 numbered leaves. (RLO-1750-83).

Tonjes, S.D. 1971. Zinc-65 uptake by a bacterium isolated from Alder Slough, Columbia RiverEstuary. Master'sThesis,Oregon State University. 38 numbered leaves. (RLO-2227-T12-18). VanArsdale,H.A. 1967. The distribution of Hyperiid amphipods off the Oregoncoast.Master's Thesis, Oregon StateUniversity. 34 numbered leaves.

Vanderploeg,H.A. 1972.The dynamics of zinc-65 in benthic fishes and their prey offOregon. Ph.D.Thesis,Oregon StateUniversity. 103 numbered leaves. (RLO-2227-Tl2-37). 314

PROGRESS REPORTS (Arranged Chronologically)

Osterberg,C.L. 1964. Radioanalysis of Oceanic Organisms in the Pacific Ocean off Oregon.Progress Report to AEC, 1 March 1963 through 29 February 1964.Reference 64-2, Department of Oceanography, OSU, Corvallis,Oregon.

Osterberg,C.L., ed. 1965. Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent PacificOcean. Progress Report toAEC,1 March 1964 through 1 July 1965.Reference 65-14, Department ofOceanography,OSU,Corvallis,Oregon.

McCauley,J.E, ed. 1966. Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent PacificOcean. Progress Report to AEC, 1 July 1965 through 30 June1966. Reference 66-8, Department ofOceanography,OSU,Corvallis,Oregon. (RLO-1750-8).

McCauley,J.E., ed. 1967. Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent Pacific Ocean.Progress Report toAEC,1 July 1966 through 30 June 196.7.Reference 67-7, Department ofOceanography,OSU,Corvallis,Oregon. (RLO-1750-22).

McCauley,J.E.,ed. 1968. Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent PacificOcean. Progress Report to AEC, 1 July 1967 through 30 June1968. Reference 68-7, Department ofOceanography,OSU,Corvallis,Oregon. (RLO-1750-28).

McCauley,J.E., ed. 1969. Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent Pacific Ocean. Progress Report to AEC, 1 July 1968 through 30 June1969. Reference 69-9, Department ofOceanography,OSU,Corvallis,Oregon. (RIO-1750-54).

McCauley,J.E., ed. 1970.Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent PacificOcean. Progress Report toAEC,1 July 1969 through 30 June1970. Reference70-20,Department ofOceanography,OSU,Corvallis,Oregon. (RLO-1750-66).

McCauley,J.E., ed. 1971.Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent PacificOcean. Progress Report to AEC, 1 July 1970 through 30 June1971. Reference71-18,Department ofOceanography,OSU, Corvallis, Oregon. (RLO-2227-T12-10).

Cutshall,N.H., ed. 1972. Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent Pacific Ocean. Progress Report toAEC,1 July 1971 through 30 June, 1972.Reference 72-19, Department ofOceanography,OSU,Corvallis,Oregon. (RLO-2227-Tl2-30). 315

McMechan,K.J., ed. 1974. Ecological Studies of Radioactivity in the Columbia River Estuary and Adjacent PacificOcean. Progress Report toAEC,1 July 1972 through 31 March 1974.Reference 74-4, School ofOceanography,OSU, Corvallis, Oregon. (RLO-2227-T12-43).